Variation in patients’ perceptions of elective percutaneous coronary intervention in stable coronary artery disease
A cross sectional study
Variation in patients’ perceptions of elective percutaneous coronary intervention in stable coronary artery disease: cross sectional study by Faraz Kureshi and colleagues (BMJ 2014;349:g5309)
Objectives—To assess the perceptions of patients with stable coronary artery disease of the urgency and benefits of elective percutaneous coronary intervention and to examine how they vary across centers and by providers.
Design—Cross sectional study.
Setting—10 US academic and community hospitals performing percutaneous coronary interventions between 2009 and 2011.
Participants—991 patients with stable coronary artery disease undergoing elective percutaneous coronary intervention.
Main outcome measures—Patients’ perceptions of the urgency and benefits of percutaneous coronary intervention, assessed by interview. Multilevel hierarchical logistic regression models examined the variation in patients’ understanding across centers and operators after adjusting for patient characteristics, using median odds ratios.
Results—The most common reported benefits from percutaneous coronary intervention were to extend life (90%, n=892; site range 80-97%) and to prevent future heart attacks (88%, n=872; site range 79-97%). Although nearly two thirds of patients (n=661) reported improvement of symptoms as a benefit of percutaneous coronary intervention (site range 52-87%), only 1% (n=9) identified this as the only benefit. Substantial variability was noted in the ways informed consent was obtained at each site. After adjustment for patient and operator characteristics, the median odds ratios showed significant variation in patients’ perceptions of percutaneous coronary intervention across sites (range 1.4-3.1) but not across operators within a site.
Conclusion—Patients have a poor understanding of the benefits of elective percutaneous coronary intervention, with significant variation across sites. No sites had a high proportion of patients accurately understanding the benefits. Coupled with the wide variability in the ways in which hospitals obtain informed consent, these findings suggest that hospital level interventions into the structure and processes of obtaining informed consent for percutaneous coronary intervention might improve patient comprehension and understanding.
Why do the study?
People with stable coronary artery disease can reasonably expect a percutaneous coronary intervention (PCI) to improve their angina. But there’s no evidence that PCI can prevent heart attacks or prolong survival
We know that stable patients overestimate the benefits of PCI. The authors of the study wanted to find out more about the extent of the misunderstanding, and begin to explore why. We don’t get all the answers, but we do get some valuable pointers about where the misunderstandings might be coming from.
Knowing what patients expect from treatment is an important aspect of shared care. If we don’t know what patients think, we can’t correct mistakes or improve education and consent procedures. No one should agree to an invasive and risky procedure without knowing what the treatment is meant to do for them, and importantly, what it is not meant to do.
What did the authors do?
The authors interviewed adults with stable coronary artery disease just after an elective percutaneous coronary intervention, asking specifically about what kind of procedure they had had (elective or urgent) and the benefits they expected to gain. The adults were given a list of options including preventing heart attacks, extending survival, improving symptoms, and improving test results. They could tick as many as they liked.
The authors started with consecutive patients scheduled for PCI in US hospitals over a two year period, and after exclusions interviewed 991 patients, treated by 135 interventional cardiologists, at 10 hospitals. They wanted to quantify patients’ expectations, quantify any variation in their expectations, then explore whether the source of the variation came from different patients, different doctors, or different hospitals.
The primary aim was to look for variation among hospitals, after controlling for patient and doctor characteristics that might influence patients’ understanding. In other words, would two similar patients treated by two similar doctors have different expectations if they were treated in different hospitals?
The authors identified important patient characteristics (such as sex, socioeconomic status, and severity of symptoms) at interview and identified doctors’ characteristics (such as age, experience, and professional certification) from publicly accessible sources, such as licensing websites, and through direct contact with their employer. They used statistical modeling techniques to look for variations in patient expectations among hospitals and among different doctors at the same hospital. They reported variation as median odds ratios, an unusual metric not to be confused with regular odds ratios, which are much more widespread in observational research. The authors also collected information about consent procedures used at each hospital, and compared them in descriptive way (without statistical tests)
What did they find?
While two thirds of patients interviewed believed that PCI could improve their angina just 1% (9/991) of patients correctly understood that improvement in symptoms was the only benefit associated with PCI. So 99% of patients had erroneous expectations. Most patients selected three, four, or five potential benefits. 90% (892/991, site range 80% to 97%) thought PCI could prolong survival, and 69% (31% to 85%) thought it could save lives. 88% thought PCI could help prevent heart attacks (872/991, 79% to 97%). One in five patients mistakenly thought their PCI was urgent, not elective. 1
The authors report significant variation in patients’ responses among different hospitals. For example, the authors report a median odds ratio of 1.72 (1.29 to 2.94) for “prevent myocardial infarction.” In other words, for two similar patients treated by similar doctors in different hospitals, one has 72% greater odds than the other of believing that PCI could help prevent a heart attack. The variability was even higher for “save life” (2.04, 1.57 to 3.54). 2
Consent procedures also varied among hospitals, including variations in the person obtaining consent, the setting, and the use of educational material. Importantly, patients’ expectations did not vary significantly among different doctors at the same hospital.
What are the study’s strengths and weaknesses?
The authors recruited a large sample of consecutive eligible patients from multiple hospitals in the US, to improve the power and precision of their estimates, compared with previous smaller studies. They achieved an excellent response rate (98.7%). For most research designs, bigger is definitely better. Bigger samples have more power to identify or rule out important differences, and generate more stable and reliable results.
Did these authors have enough power to rule out significant variation among doctors in the same hospital? It’s hard to say. Results that declare “no significant difference” should always be treated with caution when you don’t have a power calculation.
Despite the authors’ efforts to recruit from multiple hospitals across the US, their participants were mostly white, male, insured, and reasonably well educated. Women and disadvantaged groups are under-represented, so we shouldn’t assume the results apply to them. Surveys of knowledge, attitudes, and beliefs can be very context specific so it might not be wise to generalize beyond the US.
The authors had limited information about consent procedures used, and no information about what doctors think of PCI for people with stable coronary artery disease. We don’t know what patients were told, or precisely where their misconceptions came from.
What do the findings mean?
US adults with stable coronary artery disease have very poor understanding of the potential benefits of PCI, which are limited to symptom control. PCI does not save lives or prevent heart attacks and patients need to understand that before submitting to an invasive procedure and associated drug treatments, including anticoagulants. Almost all the patients in this study were expecting too much from their PCI and some put themselves in harm’s way as a result by accepting a treatment they would otherwise have declined.
Why are misconceptions so widespread in the US? We don’t yet know but this study gives us a clue that hospital factors, perhaps consent procedures, justify closer scrutiny in future research. There seem to be substantial variations in the level of patient misunderstanding among hospitals, and these variations are not explained by measured characteristics of either patients or doctors.
These authors weren’t able to identify the hospital factors linked to more or less misunderstanding among patients, but they suspect that organizational culture could be important. Future research should consider whether a hospital culture that fosters openness, good communication, patient partnership, and shared decision making is associated with better informed patients and doctors.Alison Tonks, associate editor, The BMJ
Correspondence to: firstname.lastname@example.org
Competing interests: AT helps select research for publication in The BMJ.
Provenance and peer review: Commissioned; not externally peer reviewed.
Cite this as: Student BMJ 2014;22:g5786