PCORI funded the Pilot Projects to explore how to conduct and use patient-centered outcomes research in ways that can better serve patients and the healthcare community. Learn more.

Background

People who have irregular heartbeats often must decide whether to have surgery to put a device called an implantable cardioverter defibrillator (ICD) in their chest or stomach area. If their heartbeat becomes dangerously irregular, an ICD would give the heart a small shock to make it beat regularly. However, ICDs can have disadvantages. An ICD might not always work when it is supposed to, or it might shock the heart at the wrong time. When an ICD delivers a shock, some patients feel frightened, anxious, or depressed.

Patients who might be helped by an ICD need to talk with their doctors about the pros and cons of having an ICD so they can choose what is best for them. Doctors and patients can use tools called decision aids to help them make decisions together. There are decision aids for many health topics. Decision aids may be printed booklets, videos, or websites. However, little research has looked at what kind of decision aid works best for patients considering an ICD.

Project Purpose

The research team wanted to find out which type of decision aid works best for people making decisions about whether to get an ICD. The team developed four versions of a decision aid: a worksheet, a paper pamphlet, a video, and a website. The researchers wanted to know which versions of the decision aid

• Were easy to use
• Were best for helping patients make decisions that were consistent with their values and didn’t cause regret or stress about the decision
• Were most likely to be used by patients

Methods

To create the four versions of the decision aid, the researchers used information from interviews with 28 patients who had been referred for an ICD or had received one. The patients were from medical clinics at three healthcare systems in the Denver area.

The researchers tested the four versions of the decision aid with 21 patients who were considering getting an ICD. The researchers randomly chose which patients got the decision aids and which did not. The team mailed 15 of the patients all four versions of the decision aid. Six patients didn’t receive a decision aid, so that the researchers could compare patients’ use of the decision aids with the usual way that patients made decisions. The researchers told patients who received decision aids that they could use them at their next doctor visits.

After the patients used the decision aids, the researchers interviewed them in person or on the phone. The patients also filled out surveys.

Findings

The researchers found that six out of nine patients thought the decision aids presented information fairly, without favoring one decision over another; two out of nine thought the decision aids favored getting an ICD; and one thought the decision aids favored not getting an ICD.

Most patients (eight of nine) found using decision aids to be helpful. All of the patients who received decision aids said they would recommend the decision aids to other patients.

The study found that patients who used the decision aids were somewhat more likely to make a decision about getting an ICD that matched their health goals than those who didn’t use a decision aid.

Overall, patients liked getting decision aids in the mail so that they had time to review the information before their next doctor visits. But some patients said they would rather get the decision aids from clinic staff. Patients preferred the pamphlet and video versions of the decision aids to the website and worksheet.

Patients said they used the decision aids in different ways. The aids helped some of them think of questions to ask their doctors. Others said decision aids made them feel better about the choices they made. Some patients used the decision aids to educate their families about ICDs.

Limitations

The study did not include many people. To measure differences in the way patients used the four versions of the decision aid, the researchers would need to include more people. The study took place at clinics in the Denver area only. The results might be different in other places.

Conclusions

Patients found that using decision aids was helpful when deciding whether to get an ICD. Patients said they would recommend decision aids to other patients.

Sharing the Results

The research team presented their findings at a national meeting and is writing articles for publication in journals (see below).

PCORI funded the Pilot Projects to explore how to conduct and use patient-centered outcomes research in ways that can better serve patients and the healthcare community. Learn more.

Background

Implantable cardioverter defibrillators (ICDs) are an ideal model of high-risk decisions as they are common (~200,000 per year) and involve complex trade-offs. Primary prevention ICDs confer a large (5 to 7 percent) absolute risk reduction in all-cause mortality over five years. Yet ICDs do not improve quality of life, and patients living with ICDs are at risk for device failure, inappropriate shocks, and shocks at the end of life. If shocked, patients experience increases in anxiety, depression, and post-traumatic stress disorder. Most patients see prevention of sudden death as desirable but also prefer to die peacefully in their sleep. In recognition of the personal nature of decisions related to ICDs and in response to poor patient understanding, professional guidelines recommend shared decision making for eligible patients.

Project Purpose

The overarching purpose of this research was to explore the optimal strategy to accomplish patient-centered care for high-risk decisions, using ICDs as a model. Patient decision aids (PtDAs) exist for varying afflictions and come in many formats (e.g., print, video, web). However, little comparative work has been done to determine the optimal design. This project aimed to (1) develop four PtDAs informed by stakeholder input, International PtDA Standards, and prescriptive theories of decision making, and (2) test the PtDAs using a randomized trial in samples of patients considering ICDs at three diverse healthcare systems.

Study Design

Formative development (Phase 1) and pilot feasibility evaluation (Phase 2) of a suite of PtDAs in varying formats using a mixed-methods evaluation

Participants, Interventions, Settings, and Outcomes

Patients were eligible for the study if they had received an ICD (Phase 1) or were referred to an electrophysiologist for an ICD (Phase 2) for primary prevention (Heart Failure Class II or III and left ventricular ejection fraction <35 percent). The study used two sampling strategies: (1) clinic staff identified eligible patients and (2) research staff reviewed electrophysiologists’ schedules for referrals. During Phase 1, the research team recruited 28 eligible patients to participate in qualitative interviews. During Phase 2, the research team recruited 21 eligible patients and enrolled and randomized them 2:1 to the intervention for the purpose of gathering more feasibility data; 15 patients were randomized to the intervention and 6 patients to the control (usual care).

Patients were recruited from three clinical settings across the Denver area (the Denver VA Medical Center, Kaiser Permanente, and the University of Colorado). Patients were recruited and interviewed either in person or by telephone. During the Phase 2 pilot trial, if patients were randomized to the intervention group, they were mailed the decision tools. Follow-up interviews were conducted over the phone.

The suite of PtDAs developed included: (1) option grid, (2) paper pamphlet, (3) video, and (4) interactive website.

There were three categories of outcomes:

1. Acceptability and feasibility: Acceptability was measured with a commonly used instrument: exploring constructs such as bias and satisfaction. Feasibility was measured through the pragmatically designed pilot trial exploring PtDA deliverability.
2. Quantitative outcomes: The quantitative outcomes included decision quality (for example, knowledge, value concordance), decision conflict, and decision regret.
3. Qualitative outcomes: The qualitative portion of the analysis provided us information in three domains: (1) process of dissemination, (2) tool preference, and (3) decision aid use.

Data Analysis

Qualitative data analysis drew on semistructured in-person or telephone interviews. Interviews addressed acceptability and usability of the decision aids, and the study team analyzed the data for themes related to decision influence. The team compared pilot trial quantitative outcomes using independent sample t tests for continuous outcomes.

Findings

Acceptability and feasibility outcomes: Sixty-seven percent of participants found the PtDAs to be unbiased, 22 percent thought the PtDAs were biased toward choosing an ICD, and 11 percent thought the PtDAs were biased toward not getting an ICD. Furthermore, 89 percent of participants found the PtDAs helpful, and 100 percent of participants would recommend the PtDAs to other patients. The pilot was feasible at all sites; however, clinic staff identifying eligible patients was more efficient than research staff identifying eligible patients through chart review.

Quantitative outcomes: This study was an underpowered pilot trial designed to prepare for the next trial. Intervention patients did not have significantly greater knowledge about ICDs at baseline (mean = 14.0, SD = 2.6) than control patients (mean = 11.6, SD = 3.1, t13 = 1.57, ns). Intervention patients showed a trend toward increased concordance between their decisions and end-of-life values (71 percent concordant versus 29 percent non-concordant, p = 0.06). Intervention patients did not have significantly different levels of decision conflict (mean = 17.8, SD = 15.0) than control patients (mean = 24.4, SD = 25.4, t13 = 0.64, ns) or decision regret (mean = 21.9, SD = 16.2, mean = 16.0, SD = 19.1, respectively, t11 = 0.59, ns).

Qualitative outcomes: The qualitative portion of the analysis provided information in three domains:

• Process of dissemination: Patients saw the use of mail as positive because there was time to review the tools prior to the appointment (although some patients would prefer getting the tools from the clinic).
• Tool preference: Receiving all four tools and then learning about their use was overwhelming to some participants. In general, participants preferred the paper version and the video. The website and the option grid were not as popular.

Tool use: Participants used the tools in many different ways. Some participants used the tools to stimulate questions, some participants used them for peace of mind, and some participants used them to inform their families.

Limitations

This study was underpowered to test for quantitative differences in constructs such as decision quality, and the study was conducted only in Denver.

Conclusions

Patients felt the PtDAs provided helpful, balanced information that they would recommend to other patients. Utilizing clinic staff was an efficacious way to get decision aids to patients