Patients’ Attitudes and Perceptions of ICDs: Interview With Rachel Lampert, MD

In this feature interview, we speak with Dr. Rachel Lampert about newly published research on patient attitudes and perceptions regarding primary prophylaxis with implantable cardioverter-defibrillators (ICDs).¹ Dr. Lampert is an Associate Professor of Medicine at Yale University School of Medicine in New Haven, Connecticut. 

What made you pursue this research? Had you done previous research in this area? 

Although it has been well described by many large studies that ICDs save lives in different populations at risk for sudden death, ICDs are still extremely underutilized.

One study, which looked at patients with heart failure in the “Get with the Guidelines” database, found that only 40 percent of people who should have been receiving the device were actually getting them. I think there are a lot of possible reasons why this may be the case. Therefore, we did a study a couple of years ago in which we looked to see whether putting a reminder on medical charts in an outpatient setting would improve utilization of ICDs or at least improve referrals to an electrophysiologist. We hypothesized that one reason ICDs were being underutilized was because the primary cardiologist was the one that needed to recognize the patient who needed the device, but, not specializing in electrophysiology, an ICD may not have been at the forefront of the cardiologist’s mind. So we put a reminder on the chart, and we found that this did improve referrals to electrophysiologists. However, we were also surprised to find in this study that there were a lot of patients who were refusing an ICD. This was something that hadn’t previously been described — that patient preference may be one reason why the device might be underutilized. We decided to research this further and talk to patients to see if we could understand why they weren’t interested in pursuing a therapy that might be potentially lifesaving. 

How were the interview questions formulated? Give us examples of some of the questions asked.

We approached this with qualitative methodology, which is common in fields such as psychology but not so common in cardiology. In your average study the researcher has a preset hypothesis, they state what they think is going on, send out a survey, and test if their hypothesis was correct. However, in a qualitative study, the researcher doesn’t know what the reason might be, so the purpose is to try to get a sense of what the patients themselves are thinking. The patients we interviewed were those who had met the criteria for getting an ICD (i.e., low ejection fractions), and we excluded those who had major comorbidities. We probed for what their feelings were about ICDs and what their feelings were about their own conditions. For example, we asked them things such as What was your understanding about what a defibrillator does? and What do you think might be good and bad about an ICD? We asked about how they had discussed it, what their fears and concerns were, and what their doctors thought about it. We also asked questions about how they thought an ICD might affect them. Half of our participants had accepted ICD referral and half had refused. 

Tell us about some of the demographics of the 25 patients in this study. 

They were similar to an average ICD population. The mean age was about 70 years old, and about a third of the patients were women. In this particular sample they all were white, which I think had to do with the geographic area where we are located in Connecticut, which is not very racially heterogeneous. Participating were the Yale Clinic in Branford, Connecticut, and two practices in eastern Connecticut.

List the major themes in patient perceptions found in this study. 

We found five main themes that seemed to influence patient decision. The first theme, which was really the most striking, was that people who had not been interested in ICD referral seemed to not have much insight into their own medical condition. A lot of people didn’t seem to accept they were at risk for sudden cardiac death. The concept that seemed especially hard was the concept of prevention — that even though right this minute they were not noticing any change in their condition, they had indicators that they might be at risk. However, it is important to note that the purpose of qualitative research is to be hypothesis generating, and since it’s not a quantitative methodology, you cannot generate statistical significance — all you can generate is a hypothesis. So based on these findings, we hypothesized that this was an important factor driving the decision of whether or not to accept an ICD. 

A second theme that the patients talked about was the strength of referral by the physicians. Many of the patients that accepted the device said “yes, my doctor said I needed to have this now,” whereas many of the patients who refused the device said “well, my doctor doesn’t seem to care one way or the other whether I did this.” Again, although you cannot compare them statistically, we would hypothesize that the strength of the physician recommendation is a very strong factor in making the decision to implant. 

One of the things that also surprised us was that a lot of the concerns that I think doctors believe drive patient decisions — concerns such as device recalls, fear of malfunction or fear of surgery — actually did not seem to play a large role based on this qualitative study. Both groups of individuals (acceptors and refusers) mentioned that they had heard about recalls, and had concerns about surgery, but this didn’t seem to play a prominent role for either group. 

The next theme was that patients demonstrated a lot of inaccurate perceptions about the risks and impact of the ICD. However, it seems that those who accepted the device had brought it up with their physician and had a chance to talk about it, while those who didn’t accept the device either didn’t ask about it or were just left with their same inaccurate perceptions. If a quantitative study were to confirm this, then it would be very important for doctors to probe patients for what they think might be the impact of an ICD, because there may be perceptions that patients are thinking about that aren’t in fact accurate. Doctors need to find out what preconceptions their patients have in order to address them. 

The fifth theme was not surprisingly that some people had feelings that at this time in their life they didn’t want any further life-prolonging therapies, and that is of course why they didn’t want the ICD. This is an entirely reasonable approach if that is consistent with the patient’s goals of care at that time. 

Describe the differences in why the decision to implant may have been seen as an “option” versus a “prescription,” and how this affected patient perception. 

Again we can only generate hypotheses, but the people who had accepted the device described that the doctor who had suggested it said this was the next phase in their care, whereas the ones who had not accepted the device had been presented with the ICD referral as an option that could go either way. We would need to confirm this statistically with a quantitative study, but it certainly seemed to be making a difference in how people were thinking about the device. 

Discuss how age played a factor in invasive life-extending measures. 

There was not an overall statistical difference between the acceptors and the refusers. The individuals who refused ICDs were younger (67 years old), whereas the mean age of the acceptors was 70, but it wasn’t statistically significant. The group was too small to determine whether age was related to those people who talked about not wanting life-prolonging measures. 

Has any further follow-up been done or is planned? 

We do want to follow this up with a quantitative type of study, which will be the next step. 

What makes this study unique? What are the implications of this research?

The reason why it is unique is because this is the first study that directly asks the patients about their perceptions of ICDs, prior to receiving them, and this is the first study to allow us to start generating hypotheses about which of these concerns may be playing the biggest role in their decision making. This is important because ICDs can save lives, and this information can help guide physicians as far as communication with their patients. Should our hypotheses be confirmed, I think there would be a lot of implications as to how doctors should be talking to patients. One of the most important things that we really need to focus on is making sure the patients themselves understand that they are at risk for sudden death and why they are at risk for sudden death, because that insight really plays a bigger role than specific details of the ICD itself. Making sure the patients understand their own risk should be the most important part of that physician/patient communication about ICD recommendation. The second implication is that doctors really need to be clear about what they’re recommending. One of the limitations of the study is that we didn’t ask the physicians what their impression was of how strongly they recommended the device, so we can’t determine whether in fact the physicians’ recommendations differed — that is another important area of research in the future. Finally, the third implication is that it’s very important that the physicians discuss during the course of their clinic visit what their patients already think about the ICD, so that if they do have concerns that are not true, this can be addressed.  

Reference

1. Yuhas J, Mattocks K, Gravelin L, et al. Patients’ attitudes and perceptions of implantable cardioverter-defibrillators: potential barriers to appropriate primary prophylaxis. Pacing Clin Electrophysiol. 2012;35:1179-1187.