This HIPAA-compliant study comprised two prospective controlled trials that were conducted independently at two academic medical centers (UC San Diego Health [site A] and Stanford Health Care [site B]). The study received approval from the institutional review board (IRB) at each site. The handling of informed consent differed between the two sites, as described later in the Methods.

The study included adult patients (≥ 18 years old) scheduled to undergo a clinic visit with an interventional radiologist to discuss and obtain informed consent for an image-guided procedure. Potential participants were screened for the following eligibility criteria: English was listed as their preferred language in the electronic medical record, and they had not previously undergone the procedure. After the date of the clinic visit, patients completed a survey assessing understanding of the procedure being discussed as well as satisfaction with and perceptions of the consent conversation that occurred at the visit. The survey was administered before the planned procedure to minimize recall bias, similar to previous studies of PDAs [16]. Surveys were completed without the referring or consenting clinician present. Patients were informed that survey responses were anonymous and would not affect their care. Patients did not need to consent to undergo, or ultimately undergo, the procedure to participate in the study and complete the survey.

At each site, patients were allocated to receive or not to receive a PDA before the clinic visit. In addition, both sites evaluated the same primary endpoints of patient understanding of the procedure and satisfaction with the consent conversation, as reported on the postvisit surveys. Both sites also used the same post-visit survey. However, methods otherwise differed between the two sites, as described later in this section. These differences related to variations in procedural volumes and infrastructure between the two sites. For example, the study investigators had originally intended to automate randomization and data collection processes to achieve double blinding at both sites. This automation was feasible at site B but not at site A because of limited information technology infrastructure such that data needed to be collected manually at site A. Sites were permitted to collect additional data for subanalyses. Site A elected to administer an additional survey before the visits during which the consent conversations occurred, and the automation at site B enabled assessment of how often patients accessed the PDAs.

PDAs were developed for the following procedures: kidney ablation, liver ablation, lung ablation, inferior vena cava filter removal, transarterial chemoembolization (TACE), transarterial radioembolization (TARE), and uterine fibroid embolization (UFE). Two-page handouts were selected as the medium for the PDAs because of initial feedback from patient focus groups that handouts were more accessible than videos. The initial draft of each PDA was developed by reviewing current literature to convey accurate balanced information about reasonable alternatives and typical complication and success rates. The drafts were then reviewed to ensure that they met ethical and legal criteria for informed consent and were written at the CDC's recommended health literacy level of sixth through eighth grade in plain language. Drafts were reviewed by interventional radiologists across practice environments who regularly perform the procedures and by focus groups with lay people from diverse backgrounds and education levels. The final draft was sent to a graphic designer to create the PDA. Example PDAs for TACE and UFE are provided in Figures 1 and 2, respectively. For a given procedure, the same PDA was used at both sites.

The survey was adapted from the informed consent assessment tool recommended by the Agency for Healthcare Research and Quality (AHRQ) [12]. The survey was designed to be able to be completed in approximately 2 minutes. The survey began with questions related to patient demographic characteristics. The survey then presented a free-response question asking, “what procedure are you having done?” (which was used to assess patients' actual understanding of the procedure, as described later in the Methods). The survey subsequently asked a series of questions assessing patients' perceived understanding of various aspects of the procedure being discussed, answered on a 5-point Likert scale from Not at all to Extremely well. The survey next asked patients to express their agreement with a series of statements relating to satisfaction with the consent conversation, using a 5-point Likert scale from Strongly disagree to Strongly agree. Finally, the survey asked patients to assess their overall satisfaction with the discussion that they had about the procedure and the information provided using a Likert scale from 1 (not satisfied at all) to 10 (completely satisfied). The complete survey is available as Appendix 1.

At site A, patients who underwent an outpatient clinic visit between August 2020 and July 2021 for a discussion and consent conversation regarding liver ablation, inferior vena cava filter removal, TACE, TARE, or UFE were screened for the previously noted eligibility criteria. One of two investigators (M.W., a 3rd-year radiology resident, or S.S., a 4th-year medical student) contacted each eligible patient by telephone either the evening before the visit or earlier in the same day as the visit. In this call, the investigator described the study and provided the patients with the option to participate or to decline. The IRB at site A waived the requirement to document written informed consent for study participation given the very low risk of study activities. Patients who agreed to participate were assigned in an alternating fashion to receive or not receive a PDA. If the patient was assigned to receive a PDA, then, during the call, the investigator sent the patient an e-mail that included as an attachment the PDA relevant to the procedure to be discussed and confirmed its receipt. During the call, the investigator also administered to all participating patients, whether the patient was assigned to receive a PDA or not, a previsit survey that contained a portion of questions from the postvisit survey (the questions relating to actual and perceived understanding of the procedure to be discussed at the clinic visit). Patients assigned to receive a PDA completed the previsit survey before receipt of the PDA. Finally, the investigator called all patients again 1–3 days after the clinic visit to administer the full postvisit survey. The research assistant recorded patient responses to the previsit and postvisit surveys during the calls. Interventional radiologists were blinded during consent conversations as to whether patients were participating in the study and whether study participants had received a PDA.

At site B, participants who underwent an outpatient clinic visit between January 2021 and October 2021 to discuss kidney ablation, liver ablation, lung ablation, TACE, or TARE were screened for eligibility. A module for administration of study activities was built within the institution's online and mobile health care application (MyHealth, Stanford Health Care) that is used by patients to schedule appointments and to access their health information. Per the institution, greater than 85% of its patient population regularly uses this application. During the study period, whenever a potentially eligible patient scheduled an outpatient clinic visit to discuss any of the included procedures, an investigator (E.J.K., a 4th-year interventional radiology resident) entered the patient's identifier into the module, along with the procedure to be discussed. The module then automatically screened the patient for the eligibility criteria according to information within the electronic medical record. The module automatically randomized eligible patients to be sent or not sent, before the clinic visit, an electronic communication through the application's online patient portal that included a link to access the PDA through the portal. The communication was sent 7 days before the visit if the patient was entered into the module more than 7 days in advance and was sent immediately on module entry if the patient was entered within 7 days of the visit. The module recorded each time that patients opened the PDA through the online portal. Interventional radiologists were blinded during consent conversations to whether patients had received a PDA. The day after the clinic visit, the module sent through the portal an electronic communication to all patients, regardless of whether they had received the PDA. This message described the study, including the earlier randomization to receive or not receive a PDA, and asked whether the patient would be willing to participate by completing an online survey, which was accessed by following a link in the electronic communication. Patients who did not follow the link to take the survey or who only partially completed the survey were excluded from the analysis. No previsit survey was administered at site B.

The IRB at site B waived the requirement for documentation of written informed consent for study participation. The post-visit e-mail communication stated that following the link to the survey indicated consent to participate, and patients provided consent only for survey completion. The IRB did not require that patients receive information regarding the study before potentially receiving a PDA or that patients consent to be randomized to receive or not receive a PDA because informational handouts were in widespread use at the institution, and the PDA was not considered to be a deviation from regular clinical activities.

One of the previously noted investigators (E.J.K.) reviewed survey responses from both sites and scored the accuracy of patients' responses to the free-response question “what procedure are you having done?” using a 3-point scale (1 = inaccurate [e.g., “I don't know”]; 2 = partially accurate [e.g., “fibroid treatment” for UFE]; 3 = accurate [e.g., “fibroid embolization”]). This investigator was aware of the procedure that each patient discussed but was blinded to whether patients had received a PDA.

Survey responses were summarized as median with IQR for age, count with percentage for other demographic characteristics, and mean with SD for the remaining questions. Responses were reported for sites independently and pooled. Mann-Whitney U, Fisher exact, and chi-square tests were used to compare responses to questions relating to demographic characteristics as well as understanding of the procedure and satisfaction with the consent conversation between patients who received and did not receive a PDA. The distribution of procedures was compared between patients who did and did not receive a PDA using Fisher exact test, treating ablation-related procedures as a single category. Wilcoxon signed rank tests were used to compare pre-visit versus postvisit responses at site A. A power analysis showed that the study would need to include 17 patients per group (PDA and no PDA groups) to detect a difference between groups of 1 on a 5-point Likert scale with 80% power and an alpha of .05. Comparisons were considered statistically significant at a p value less than .05. All statistical analyses were performed using RStudio 1.3.1056 (RStudio, PBC).

At site A, approximately 100 patients were scheduled for an outpatient clinic visit to discuss one of the included image-guided procedures during the study period and were thus screened for study eligibility. Of these, 55 were eligible, and attempts were made to contact them by telephone before their clinic visit. Four eligible patients could not be reached by telephone before their clinic visit. The remaining 51 patients were reached by telephone before the visit. All patients who were reached by telephone agreed to participate. Twenty-seven patients (14 men, 13 women; median age, 65 years) were allocated to receive a PDA, and 24 patients (13 men, 11 women; median age, 67 years) were allocated to not receive a PDA before the clinic visit. All of these patients completed the surveys administered by telephone both before and after the clinic visit, for a 100% (51/51) survey completion rate. These 51 patients formed the final study sample from site A.

At site B, approximately 120 patients were scheduled for an outpatient clinic visit to discuss one of the included image-guided procedures during the study period and were thus screened for study eligibility. Of these, 92 were eligible and were randomized to receive or not receive a PDA before the clinic visit. After the clinic visit, all 92 patients received an e-mail communication inviting them to complete the survey. Thirty patients either declined to take the survey or only partially completed the survey and were thus excluded. The remaining 62 patients completed the survey, for a 67% (62/92) survey completion rate. Of the 62 patients who completed the survey, an additional eight patients were subsequently excluded from the analysis because later manual record review indicated that they had a preferred language other than English or had previously undergone the procedure, despite being deemed eligible by the earlier automated screening process conducted by the module. The remaining 54 patients formed the final study sample from site B. Of these patients, 26 (15 men, 11 women; median age, 70 years) had been randomized to receive a PDA, and 28 (17 men, 11 women; median age, 68 years) had been randomized to not receive a PDA.

Figure 3 summarizes the flow of patient selection at the two sites.

Table 1 summarizes demographic characteristics and procedures discussed for study participants. Between the two sites, a total of 105 patients (59 men, 46 women; median age, 67 years) were included; 53 received a PDA, and 52 did not receive a PDA. For the two sites separately and pooled, there were no significant differences between patients who received and did not receive a PDA in terms of age, sex, race and ethnicity, or distribution of procedures discussed (all p > .05).

Table 2 summarizes results of survey questions regarding actual and perceived understanding of the procedures being discussed. According to the classification of patients' free-text descriptions, patients who received a PDA, compared with those who did not, showed greater accuracy of describing the procedure (2.6 vs 2.3, p = .04). In addition, patients who received a PDA, compared with those who did not, reported greater self-perceived understanding of what the procedure involved (4.4 vs 4.1, p = .02), as well as of the procedure's benefits (4.5 vs 4.0, p < .001), risks (4.4 vs 3.6, p < .001), and alternatives (4.0 vs 3.3, p < .001).

Similar findings were observed separately for sites A and B with a small number of exceptions. At site A, actual accuracy of procedure descriptions was not significantly different between patients who did and did not receive a PDA (2.6 vs 2.5, p = .44). At site B, patient's self-perceived understanding of what the procedure involved was not significantly different between patients who did and did not receive a PDA (4.3 vs 3.9, p = .06).

Table 3 summarizes results of survey questions regarding satisfaction and perceptions with the consent conversations. Patients who received a PDA, compared with those who did not, were more likely to feel that their doctor spent enough time with them (4.7 vs 4.5, p = .03), listened carefully to them (4.8 vs 4.4, p < .001), provided enough time to make a decision (4.8 vs 4.3, p < .001), encouraged them to ask questions (4.8 vs 4.5, p < .001), and answered all of their questions (4.8 vs 4.4, p = .001). Patients who received a PDA, compared with those who did not, were also more likely to feel free to choose any option, including not to have the procedure, than patients who did not receive a PDA (4.7 vs 4.3, p < .001). Finally, patients who received a PDA, compared with those who did not, showed overall greater satisfaction with the discussion and the information provided, although this difference was not statistically significant (9.6 vs 9.3, p = .06).

Similar findings were observed separately for sites A and B with a small number of exceptions. At site B, patient's likelihood of feeling that the doctor spent enough time was not significantly different between patients who did and did not receive a PDA (4.7 vs 4.5, p = .23). At site B, overall satisfaction was significantly higher for patients who did than did not receive a PDA (9.5 vs 9.0, p = .03).

Table 4 summarizes results of previsit and postvisit survey questions regarding understanding of procedures at site A. When comparing previsit and postvisit results, patients in both groups showed an increase in actual accuracy of descriptions of the procedure being discussed (PDA, 1.6 vs 2.6; no PDA, 1.6 vs 2.5), as well as increases in self-perceived understanding of what the procedures involved (PDA, 2.6 vs 4.5; no PDA, 2.6 vs 4.3), and its risks (PDA, 1.8 vs 4.3; no PDA, 2.0 vs 3.8), benefits (PDA, 3.4 vs 4.6; no PDA, 2.8 vs 4.1), and alternatives (PDA, 1.9 vs 4.0; no PDA, 1.8 vs 3.6) (all p < .001). On the previsit survey, patients who did not receive a PDA had the highest scores for understanding of the procedure's benefits (3.4) and what it involved (2.6) and lowest scores for understanding of the procedure's risks (1.8) and alternatives (1.9).

At site B, all participants randomized to receive a PDA opened the document at least once. The number of times that the document was opened ranged from one to seven times for ablation procedures, one to three times for TACE, and one to five times for TARE.

____ years old