Perceptions of patients and providers on myocardial perfusion imaging for asymptomatic patients, choosing wisely, and professional liability

Waste in medical spending is a widespread problem, estimated to be 30% of all health spending and equivalent to 750 billion US dollars annually [1, 2]. A variety of initiatives have been created by the medical profession to curb unnecessary spending, such as the Choosing Wisely (CW) campaign. This initiative was started by the American Board of Internal Medicine Foundation (ABIM) and asked medical societies to identify five things that patients and physicians should question. Lists created by multiple specialty societies cite overuse of nuclear cardiac stress testing or myocardial perfusion imaging (MPI) [3]. High rates of unnecessary MPI testing places burden on the medical system while exposing patients to unwarranted risks. In another initiative to reduce unnecessary and wasteful testing, professional societies have also developed Appropriate Use Criteria (AUC) to assist doctors in deciding which patients may benefit from MPI in common clinical scenarios [4].

These efforts are laudable, but their acceptance and implementation by front line providers in clinical care is unknown. Furthermore, the role of patients in these processes is limited, or nonexistent; patients’ reflections on appropriateness and awareness of waste and liability risks have not been studied. To investigate these issues, we conducted a survey of patients and healthcare providers asking a variety of questions related to these issues. We hypothesized that patients would be more likely than providers to endorse the use of MPI in the absence of symptoms.

The survey was completed by 456 respondents, 342 patients and 114 providers. Three hundred fourty two patient respondents were distributed equally between cardiology (32.2%), primary care (36.0%), and nuclear medicine (31.9%) clinics; 55.6% were Veterans. (Table 1) Patient respondents were predominantly male (68%), and finished at least some college education (67.5%). Providers were from academic practice (61%), VA practice (27%), or private practice (10%). The majority were physicians (87%) followed by advanced care providers (13%). Primary specialty was cardiology (35.1%), primary care (37.7%), hospital medicine (22.8%), and other (4.4%).

We performed a large survey of both patients and providers taken predominantly from a single academic health center and affiliated VA hospital system to gauge opinions and attitudes about MPI, appropriateness, and a variety of related topics. The results of the survey indicate areas for improvement in understanding about MPI and provide some striking observations about unnecessary testing and professional liability.

Patient respondents to our survey were more likely to report that MPI was warranted in a variety of clinical scenarios for asymptomatic patients. This is in accord with a recent comprehensive systematic review that showed patients tend towards overestimating benefits while underestimating harm regardless of the intervention [9]. Patients are also poorly informed by their providers on the risks of over-diagnosis and overtreatment, as shown in a survey which noted only 10% of individuals were explained these risks when undergoing cancer screenings [10]. Contributing to information mismatch is the fact that providers are more likely to explain the pros of testing and treatments instead of the potential cons, and patients often do not feel involved in shared decision making [11‐13]. Informed-decision making strategies via balanced discussions between patients and providers are key to improving these disparities.

In free-text responses about inappropriate tests, both providers and patients offered many accurate reasons why testing is unnecessary. Providers gave responses that matched a number of the AUC for MPI. Both also gave responses that indicated ongoing concerns and awareness of professional liability. Not surprisingly, our data contributes to the literature that physicians are less concerned about professional liability when they are employed by an entity with sovereign immunity [14, 15]. We are concerned by the observation that over one third of providers openly admitted to ordering MPI specifically due to professional liability concerns. Similar findings were reported in a 2011 survey of primary care physicians showed that 76% ordered more aggressive testing due to liability concerns. [16] Another recent survey by an independent group for the American Board of Internal Medicine Foundation Choosing Wisely campaign reported that 52% of physicians felt the major reason for ordering unnecessary testing was professional liability concerns [17]. We observed among narrative responses from providers, liability concerns was the third most common “inappropriate” indication for ordering MPI behind asymptomatic patients and patient/family demand. In the same CW survey, 53% of ordering providers stated they would order an unnecessary test if a patient was insistent compared to 31% in our survey. Recent data suggest that “demanding” patients, those who go a clinical encounter with a specific request for a test or procedure, typically uncommon. Only 9% of encounters in outpatient oncology centers had such patients and that typically providers agreed with the patients’ concerns [18]. Only 1% of these encounters were deemed to have a clinically inappropriate patient demand and that these were rarely fulfilled. While patient satisfaction appears to improve with increasing knowledge about their conditions, medications and procedures, some tests and procedures may be limited in their opportunity for patient involvement in decision-making [19]. The core of the CW campaign and patient-centered medicine is to respond to “demanding” patients with respectful understanding and communication about their concerns [20].

While a proportion of providers in our survey were aware of CW (37.2%), we noted this was higher than national averages based on CW random polling data (21%) [17]. We suspect this is secondary to our study being performed at an academic medical institution. Awareness of the campaign among patients was only 2.7% in our study as opposed to about 5% as a national average. This demonstrates a continued disparity between patient and provider knowledge of the campaign and low awareness of the campaign overall.

Respondents to our survey had modest awareness of waste in the US healthcare system and efforts to reduce it. Patients were less likely than providers to accurately estimate the amount of waste, both measured in US dollars and as a proportion of total spending. Polled physicians feel aware of costs for tests/treatments and think providers should adhere to guidelines that discourage marginally beneficial care [21]. An overwhelming 89% agreed that doctors should play a bigger part in reducing unnecessary tests. A recent survey by Colla et al. showed nearly all physicians felt a responsibility to limit unnecessary testing, control costs, and adhere to guideline recommendations (96.8%, 92.2%, and 97.9% respectively) [22]. Another important finding in our survey is an estimate of the proportion of “rarely appropriate” tests that would be acceptable. The Institute of Medicine has shown that over a quarter of the $750 billion dollar annual cost of waste in health care is directly attributed to unnecessary services [2]. Currently, there is no agreement on what an acceptable rate of unnecessary services should be, but the most common response to this in our survey suggests a goal of less than 5% would be reasonable [7].

Patient responses also indicated a poor understanding of radiation exposure with 42% of respondents stating “no radiation exposure” associated with nuclear testing and 46% reported a “slight increase of exposure”. This is similar to a recent survey showing that 85% of patients undergoing CT or cardiac SPECT imaging underestimated the amount of radiation and 88% were not worried about scan radiation [23]. Another survey noted only 3% of patients had been informed about radiation risks prior to their last imaging study [24]. This is a major concern when its estimated that greater than 10% of the ionizing radiation exposure in the U.S. can be directly attributed to nuclear stress testing [25]. Furthermore, providers are also often misinformed about radiation exposure [26]. A 2011 survey of American Society of Nuclear Cardiology members showed only 9% could accurately estimate the radiation exposure of the most common nuclear stress test protocol [27]. Real time feedback has been proposed to assist with this problems as one recent study showed that displaying radiation exposure risks and imaging test costs at the time of ordering influenced providers on test appropriateness and helped facilitate physician-patient discussions [28]. More research is necessary to better guide informed decision-making for both patients and providers on radiation risks.

Our investigation has important limitations to acknowledge. The survey was conducted at a single site and may not be representative of broader populations. The methods used for population sampling may bias results towards a more favorable view of MPI. Responses were voluntary and rate of returned surveys were not tracked per Institutional Review Board regulations, leaving potential for bias due to absent responses. This is offset by the anonymous nature of the survey, especially regarding questions about professional liability and unnecessary testing. Survey responses by providers may not represent their actual clinical decision-making and ordering habits due to social desirability bias.

While efforts such as Choosing Wisely strive to improve communication and reduce low-value care, awareness and implementation of the campaign in the community remains low. Liability concerns and patient demand may still contribute to test overuse and ordering in rarely appropriate circumstances. Providers reported low rates of using AUC and had concerns about their purported benefits. Both patients and healthcare providers underestimate the costs of wasteful medical spending, patients to a greater degree. These findings suggest that further study and resources are needed to encourage patients and providers to adopt cost-conscious and appropriate medical testing.