Effects of a Patient Activation Tool on Decision Making Between Surgery and Nonoperative Management for Pediatric Appendicitis

Key Points Question Can a patient activation tool improve decision-making and patient-centered outcomes in pediatric patients and their caregivers choosing between surgery and nonoperative management for pediatric appendicitis? Findings In this randomized clinical trial that included 200 patient-caregiver dyads, groups randomized to the standardized surgical consultation and the patient activation tool had similarly high scores on immediate decisional self-efficacy and health care satisfaction measures at discharge and no differences in disability days or the failure rate of nonoperative management at 1 year. Meaning Although a patient activation tool did not improve measures of decision making, pediatric patients and families can effectively participate in an informed shared decision-making process in the acute care setting.


Randomized Controlled Trial of a Patient Activation
Surgeons provide generalized information about treatment options to acutely ill patients and time for shared 110 decision making is limited. This is especially true in vulnerable populations such as children, in which 111 caregivers are forced to take in large amounts of information and make quick decisions for someone other than 112 themselves. We hypothesize that strategies to activate caregivers in the emergency setting through knowledge 113 and engagement may result in improved decision making with improved self-efficacy and confidence. This 114 could lead to improvements in other patient centered outcomes such as health related quality of life and 115 healthcare satisfaction for both the caregiver and the pediatric patient. In this study, we define an activated 116 patient-caregiver dyad primarily on the willingness, knowledge, engagement, and self-efficacy of the caregiver. 117 118 Our study has the potential to significantly alter patient health, well-being, and the quality of care that patients 119 receive in four ways: First, patient activation in pediatric patient-caregiver dyads will be examined. The 120 majority of studies on patient activation, decision making, and patient centered outcomes focus on autonomous 121 adult patients. 9-13 However, the role of the caregiver is crucial in all of pediatric medicine. 14 Despite this, there 122 are few studies that examine patient-caregiver dyads in the context of clinical decision making and even fewer 123 that study children. 14-20 Second, this study focuses on patient activation in the emergency care setting. Most 124 research on patient activation, decision making, and patient centered outcomes is focused on chronic illness 125 and/or outpatient care. 9-13 Third, this study investigates the effects of patient activation on decision making and 126 patient centered outcomes in a surgical disease. Several studies investigating the ability of decision aids to 127 improve patient knowledge, one component of activation, have been performed in patients undergoing major 128 elective surgery. 13 However, none of these studies targeted emergency surgical procedures or children. Results 129 from our study will provide information on the utility of a PAT to facilitate patient-caregiver dyad activation in 130 emergency surgical interventions and in children. Fourth, this study will use a multi-media patient activation 131 tool that incorporates patient activation strategies with a decision aid. While traditional paper-based tools and 132 multi-media tools may contain identical information, the use of a multi-media tool offers the advantage of 133 customization, feedback, and greater visual features. [21][22][23] In addition, multimedia has the added benefit of 134 motivating participants to learn topics that they would usually avoid and obtain information they would not seek 135 out from traditional sources. [24][25][26][27] Furthermore, the PAT incorporates interactive technology designed to promote  136  patient activation, improve decision making, and effectively communicate and resolve uncertainty, in addition  137  to eliciting rapid understanding of the disease and treatment options.  138  139 We are currently performing a pilot study offering nonoperative management as a treatment choice to patients 140 and their caregivers. Thus far, 80% of children initially treated non-operatively did not require appendectomy, 141 and no patients have had an adverse event or progressed in disease severity. We believe that both non-operative 142 management and appendectomy are safe and effective treatment options for early appendicitis. As such, we are 143 beginning to offer antibiotic therapy as an alternative treatment option to patients outside those participating in 144 the pilot study. Although we believe we have enrolled a sufficient number of patients to determine that non-145 operative management is safe, we are continuing to enroll patients in the pilot study in order to more precisely 146 determine the recurrence rate. 147 148 However, determining the best treatment requires engaging the patient and caregiver in the treatment decision. Phase 2:

204
After the development of a PAT, we will conduct a prospective, single-blind, single-site, randomized trial to 205 determine if providing patient-caregiver dyads with a PAT improves decision making and patient centered 206 outcomes without compromising medical outcomes compared to standard surgical consultation alone in 207 pediatric patients (7-17 years old) with early appendicitis and their caregivers. The patient-caregiver dyads and 208 the physician research members administering the PAT will not be blinded. All clinicians subsequently 209 interacting with these patients will be blinded as to whether the patient received the PAT or not. This will be an interactive program that will allow patients to choose an avatar who will guide them through an 261 educational program. This will be designed for both the patient (child) and caregiver, and is meant to present 262 information about their condition (acute appendicitis) and the treatment options (appendectomy or antibiotics). 263 In addition to providing information regarding appendicitis and the two treatment options, the PAT will 264 include two separate brief video clips that encourage active involvement in the treatment choice and teach 265 patient-caregiver dyads how to communicate decisional uncertainty. 266 267 Phase 2: (Target start date 3/1/14)

268
Screening and Enrollment 269 270 All children age 7-17 years with a diagnosis of early appendicitis treated at NCH will be screened for study 271 eligibility. Once eligibility is confirmed, and consent obtained, the patient-caregiver dyad is randomized to 272 either a standard surgical consultation (control arm) or to a surgical consultation plus the PAT (intervention 273 arm) by physician research members trained to perform both types of consultations. The consultations will 274 occur in the NCH Emergency Department or one of the inpatient units at NCH. They will then be consented and 275 consultation will commence. 276 277 Study Procedures: All randomized patients will be approached by a trained physician research member who will conduct the 281 consultation prior to meeting the surgeon. 282 283 Patient-caregiver dyads randomized to the standard consultation will receive only verbal information regarding 284 the two treatment options (operative or non-operative management) with patients having the opportunity to ask 285 questions. This type of consultation is akin to a typical surgical consultation in which surgeons verbally inform 286 patients of their options and elicit questions. 287 288 A trained physician research member will administer the PAT among patient-caregiver dyads randomized to the 289 PAT and then answer questions about the risks and benefits of each treatment option. If the child is in 5 th grade 290 or higher, they will have the option of having their own tablet to watch the PAT. 291 292 After the physician research member has answered all patient and caregiver questions, the patient-caregiver 293 dyad will choose either appendectomy or non-operative management. Discrepancies between the caregiver's 294 choice and the patient's choice will be recorded, but the caregiver, by law, will make the decision. Both arms 295 will complete computer-based questionnaires that measure patient and caregiver recall, caregiver activation, 296 preparedness for decision making, decision self-efficacy, and decisional conflict after the treatment choice has 297 been made. Any patient or caregiver who wishes to speak with a member of the surgical team prior to making a 298 decision will be able to do so. 299 300 The physician research member will then inform the clinical team of the treatment choice and the surgeon will 301 be called to speak with the family for a full discussion of the care plan and to obtain surgical consent if 302 operative management was chosen. 303 304 The patient will be admitted to the hospital and receive care according to standard protocols for either 305 appendectomy or non-operative management of appendicitis. Based on the results of our pilot study, we have 306 demonstrated that non-operative management is a reasonable alternative to appendectomy for cases of early, 307 acute appendicitis. As such, we are presenting both options as initial treatment options for early acute 308 appendicitis to patients and their caregivers. 309 310 Before discharge, the patient and legal guardian will complete another set of questionnaires that measure: the 311 patient's quality of life (QOL), caregiver satisfaction of care received, decision self-efficacy, decisional regret, 312 reasons for initial choice and general socioeconomic status (SES) information. We will give English-speaking 313 participants and families the option of conducting follow-ups via phone call and/or email. The emails will be sent 314 from REDCap to ensure security. 315 316 A follow-up reminder letter will be mailed/ emailed/texted to all families to reiterate follow-up time points. Depending on the child and legal guardian's preference, they will either return to NCH or a member of the 336 research team will call or email the family for a follow-up. In addition, if they have already scheduled a 337 follow-up appointment with their surgeon, we will attempt to schedule our appointment during the same 338 day. The child and legal guardian will complete another set of questionnaires that measure: the patient's 339 quality of life (QOL), caregiver satisfaction of care received, decision self-efficacy, decisional regret, 340 satisfaction with decision and patient and caregiver recall (if patient is in the 5 th grade or higher and they 341 were given their own tablet to watch the PAT). 342 343 We will also ask questions about any problems the child has had, dependent on the treatment option, with 344 their appendix or surgery since discharge, any other issues, and how many days the child missed school 345 and the legal guardian missed work and any out-of-pocket expenditures (e.g. insurance copayments, and 346 outpatient medications). This will be conducted either in person, over the phone or email. A $25 VISA® 347 ClinCard will be provided. If the child and legal guardian prefer a phone or email follow-up, the gift card 348 will be mailed to the legal guardian's address. A parking voucher will be provided if they decide to come 349 back to NCH for their 30 day follow-up. 350 351 Data will also be collected from additional chart review and claims data. This will include: appendicitis up. Any recurrence or other complications will be discussed. Also, additional missed days from school 361 and work will be assessed as well as any additional out-of-pocket expenditures (e.g. insurance 362 copayments, and outpatient medications), if any. A $30 VISA® ClinCard will be provided after successful 363 completion of the 6 month follow-up. The gift card will be mailed to the legal guardian's address. 364 365 Data will also be collected from additional chart review and claims data. This will include: appendicitis 366 recurrence with determination of simple vs. complicated appendicitis from operative and pathology 367 reports; subsequent ED visits or physician visits; subsequent readmissions and reoperations; additional 368 imaging, pharmaceutical, and laboratory procedures related to their chosen appendicitis treatment, if any. 369 370 1 year (±30 days) 371 372 A member of the research team will call or email the family and ask questions about any problems the 373 child has had, dependent on the treatment option, with their appendix or surgery since their 6-month 374 follow-up. Any recurrence or other complications will be discussed. Also, additional missed days from 375 school and work will be assessed as well as any additional out-of-pocket expenditures (e.g. insurance 376 copayments, and outpatient medications), if any. A $35 VISA® ClinCard will be provided after successful 377 completion of the 1 year follow-up. The gift card will be mailed to the legal guardian's address. 378 The child and legal guardian will also complete the PEDSQL Quality of Life Inventory: Child and Parent 380 Report, and the legal guardian will complete the Satisfaction with Decision Scale. 381 382 Data will also be collected from additional chart review and claims data. Secondary endpoints are disability days at 30 days and 6 months, readmission rate and length of stay. We also 426 hope to determine the rate of recurrence of acute appendicitis for those patients who opt to receive antibiotic 427 therapy and the rate of surgical complications for patients undergoing appendectomy. This outcome will be 428 measured at the aforementioned time points, and will also be measured on an annual basis. Patient-centered endpoints to be analyzed using mixed models across all time points at which the 433 measure was determined are total decision self-efficacy scores, total PedsQL TM healthcare satisfaction generic 434 module scores, and total PedsQL TM HRQOL general core scale scores in patients and caregivers. Time will be 435 included as a categorical variable in each of these models, and its interaction with the PAT intervention will 436 also be included. The interaction terms will provide estimates of the differences in these patient-centered 437 outcomes between PAT and standard surgical consultation arms at each time point in the study. An unstructured 438 error covariance approach will be used to account for the within-subject correlation 439 in outcomes measured over time. 440 441 For disability days in a year, we will test for non-inferiority with a margin of less than two days. It is possible 442 that the PAT will reduce disability days. The non-inferiority test can be followed by a superiority test for 443 disability days, with no effect on Type I error. The PAT and standard surgical consultation groups will also be 444 compared on their average readmission rates. We will assume a negative binomial distribution using GENMOD 445 procedure in SAS 9.3 to model the count of days. 446 447 Subgroup Analyses

448
We will perform exploratory subgroup analyses to examine outcomes that are relevant only to either the 449 operative or non-operative treatment groups. This will be especially useful if we find a difference in disability 450 days between the randomized groups, because the difference may be related to differences in the percentage of 451 patients within the PAT group compared to the standard surgical consultation group that choose operative vs. 452 non-operative management. 453 454 Within the subgroup of patients that choose operative therapy, we will compare post-operative infections, re-455 operation rate at one year, and readmission rate at one year to identify if receiving the PAT altered these 456 medical outcomes. 457 458 Within the subgroup of patients who chose non-operative therapy, we will examine the need for appendectomy 459 during initial admission, recurrence of appendicitis at one year, and antibiotic side effects at 30 days to identify 460 if receiving the PAT altered these outcomes. Proportions will be compared for antibiotic complications, 461 recurrence, post-operative infection, and complications. Average readmission rates will be compared using a 462 Pearson chi-square test. healthcare satisfaction at discharge. We expect the PAT will increase the total Decision Self-Efficacy Scale 471 score by 0.4 standard deviations, and increase total PedsQL TM healthcare satisfaction generic module scores at 472 discharge by 0.5 standard deviations. [29][30] Assuming that we will show superiority for one or both of these 473 patient centered outcomes, we also consider it critical to claim that the PAT does not increase the number of 474 disability days over a year as compared to standard surgical consultation. Based on our previous research and 475 published data, we expect the number of disability days in the standard surgical consultation group to be 14 476 days with a standard deviation of 5 days. 31 Based on this, we set our noninferiority margin to 14%, or two days 477 more. For the primary patient centered outcomes we will use Holm's procedure to test the two primary 478 outcomes. Non-inferiority of disability days will only be meaningful if we achieve significance on these 479 primary outcomes. Given this, we plan to use a gatekeeping fallback procedure, whereby we can test for non-480 inferiority at the one-sided 0.025 level when we achieve significance for one or both of the patient centered 481 primary outcomes. 32 If we fail to reach significance on these primary outcomes, then we will have to use a 482 much more stringent alpha of 0.005 (fallback). In order to use this procedure, our significance level for the 483 patient centered primary outcomes will be set at 0.02 (rather than 0.025). 484 485 Based on this strategy, 100 patients in each group will provide power of 92% to reach significance on either 486 primary patient centered outcome, and 74% power to reach significance on both, assuming the effect sizes listed 487 above. This sample number provides 80% power to claim significant non-inferiority of less than two days. There are no increased risks to the standard surgical consultation group. However, the surveys 514 used to assess decision making and patient centered outcomes may make the subjects feel 515 uncomfortable due to the inclusion of questions that ask about their emotional and social 516 functioning. 517 518 Consultation with PAT 519 Potential risks may include increases in decisional conflict or regret based on being activated and 520 receiving information that might lead to more stress about their decision or may lead them to a 521 decision they may not have made prior to viewing the PAT. In addition, the surveys used to 522 assess decision making and patient centered outcomes may make the subjects feel uncomfortable 523 due to the inclusion of questions that ask about their emotional and social functioning. 524 525 Hospital Course The inherent risks of choosing non-operative management or appendectomy. For non-operative 528 management, patients may require surgery and for appendectomy, patients may develop intra-and 529 post-operative complications. 530 531 532 In addition, the surveys used to assess decision making and patient centered outcomes may make the The potential benefits of non-operative management include avoiding surgery, less pain, and 558 quicker return to normal activities. 559 560 Surgery Group

561
The potential benefit of an appendectomy is that it is curative for appendicitis. 562 563 6.2.3. Risk-Benefit Assessment/Risk Minimization 564 565 All patients will be informed that their participation in this study is voluntary and that their medical care 566 will be unchanged whether they agree or do not agree to participate in this study. Further, they will be 567 informed that if they agree to enter this study, they may withdraw from the study at any time. 568 569 Privacy of individuals and confidentiality of data will be maintained by minimizing the amount of 570 identifiable data obtained as much as possible. Only study IDs will be used to identify patients on all 571 data forms and in all datasets used for analysis. The file linking study IDs to patient names and MRN 572 will be password protected and will not be made available to non-study staff or used during data 573 analysis. All information will be compiled and stored in REDCAP and will be password protected. Strict 574 standards of confidentiality will be upheld at all times. 575 576 All patients enrolled in the trial will be monitored according to the standard care protocol for their chosen treatment option (operative or non-operative management). Any adverse events will be handled 578 according to hospital procedure. Before discharge, patients and their caregivers will be given clear 579 instructions on taking either oral antibiotics or pain medicines and for reporting any concerns while on 580 these medications. A research phone number will be provided in the informed consent should they have 581 any questions or concerns relating to the study and contact information for medical questions will be 582 provided consistent with the standard of care for surgical discharges at our institution. After hospital 583 discharge, any further complications will also be assessed. Participants may call the research staff at any 584 time to address concerns and a physician member of the research team is available to provide medical 585 advice. This will be stated in the informed consent. 586 587 Regarding the patient centered surveys, participants have the option of skipping questions they find to be 588 uncomfortable or simply wish to refuse answering. This will be stated on the informed consent 589 590 6.2.4. Data safety and monitoring 591 592 This study is only minimal risk. Data will be monitored by research team members once a week. They 593 will ensure that all data (clinical data and questionnaires) collected are correctly completed. The clinical 594 study team will review adverse events as they occur. 595 596 597 6.2.5. Adverse events (AE) and Serious adverse events (SAEs) 598 599 Adverse events which may occur during this study include increases in decisional conflict or regret 600 based on being activated and receiving information that might lead to more stress about their decision or 601 may lead them to a decision they may not have made prior to viewing the PAT. In addition, the surveys 602 used to assess decision making and patient centered outcomes may make the subjects feel uncomfortable 603 due to the inclusion of questions that ask about their emotional and social functioning. Several adverse 604 events may occur as part of the treatment choice that are not related to enrollment in the trial: All 605 patients could have adverse reactions to the antibiotics; patients who choose non-operative management 606 could have interval progression of appendicitis or the need to be readmitted due to recurrence of disease; 607 and patients that choose operative management could have intra-operative or post-operative 608 complications such as bleeding and abscesses. These are all expected complications that will not be 609 reported to the IRB. 610 611 We do not expect any severe AE in either study arm. All unexpected non-serious adverse events and 612 serious adverse events relating to the administration of the PAT will be reported verbally and in writing 613 to PCORI and the NCH IRB chair. The verbal report will occur within 48 hours of the occurrence. The 614 written report of the serious adverse event (e.g., death or life-threatening adverse event) will be reported 615 within seven days. 616 617 618 6.3. Recruitment Strategy 619 620 No active recruitment will take place. Subjects will be screened for eligibility once a diagnosis of early 621 appendicitis has been determined and the surgical consult team has been contacted. 622 623 Eligible patients are identified by the emergency department (ED), who is typically responsible for the initial