Patient perceptions of diagnostic certainty at discharge and patient satisfaction in the emergency department.

Abstract

Patient satisfaction has evolved into a standard measure for quality and value in health care. Given the importance of patient satisfaction to overall hospital quality measures, a growing literature has investigated a number of variables that affect satisfaction in the emergency department (ED). For instance, studies have demonstrated that certain objective visit-related metrics, such as reduced wait time to see a provider, shorter length of stay, and a higher number of administered treatments, underlie higher patient satisfaction in the ED.1-3 However, recent research has also highlighted that subjective measures of a patient's experience may be greater determinants of satisfaction than objective measures of care.4 For example, patient perception of wait time is a stronger predictor of satisfaction than objectively measured wait time.5, 6 These results reveal the importance of understanding the subjective aspects of a patient's experience in an ED that may predict overall satisfaction. We investigated one critical factor that may influence patients’ subjective experience (and, in turn, satisfaction): perceived diagnostic certainty—or the extent to which patients leave the ED feeling certain they know what caused their underlying condition. Psychology research has established individuals’ aversion to uncertainty and ambiguity and the negative affect induced by decisions involving uncertainty.7 There is reason to believe that diagnostic certainty may also play a role in patient satisfaction within the ED. Given the nature of the ED practice environment (i.e., relatively limited information, little to no prior relationship with patients, time constraints), the main goal is often to rule out truly emergent causes of patients’ presenting complaints—not necessarily to achieve a definitive diagnosis. Yet, many patients present with expectations of receiving a definitive diagnosis.8 Violating this expectation by discharging patients without a definitive diagnosis, thus leaving them in an otherwise uncomfortable state of uncertainty, may decrease positive affect and, in turn, decrease patient satisfaction. To develop a baseline understanding of this phenomenon, we sought to (1) determine whether perceptions of diagnostic certainty are associated with greater patient satisfaction and (2) assess the strength of the relationship between diagnostic certainty and patient satisfaction relative to other potentially relevant variables, including reported level of pain, length of stay, and number and types of follow-up recommendations provided. We administered a survey to a convenience sample of patients in a single academic tertiary care ED with an annual volume of 55,000 visits. Patients with a chief complaint of abdominal pain, back pain, chest pain, or headache who were listed for discharge from the ED were identified and approached by a research assistant between 8:00 a.m. and 11:00 p.m. These conditions were chosen by consensus because they were thought to reflect conditions where the focus of the emergency physician is often ruling out the worst-case scenario and where a definitive diagnosis might often not be reached even after a thorough ED evaluation. A survey was created using online software and given to each patient for completion on an electronic tablet. Each patient survey was paired with data obtained from an administrative database that provided details about the patient's stay, including length of stay, tests ordered, and medications administered. All survey items and data collection methods were approved by the institution's ethical review board. Our primary measures were patient affect, perceived diagnostic certainty, and satisfaction with care. To measure affect, participants rated (using a modified 7-point Likert scale ranging from 1 = not at all to 7 = very much) the extent to which they felt anxious (reverse-coded); relieved, happy, satisfied, angry (reverse-coded); frustrated (reverse-coded); nervous (reverse-coded); discouraged (reverse-coded), and confused (reverse-coded) based on their visit. We averaged all 10 affect items to create a composite of positive affect (α = 0.87). To measure perceived diagnostic certainty, participants indicated their level of agreement (1 = strongly disagree, 7 = strongly agree) with two statements: I am sure about exactly what is wrong and My doctors know exactly what is wrong. We averaged the two measures to create a composite of diagnostic certainty (α = 0.86). Finally, to measure satisfaction, patients indicated their level of agreement (1 = strongly disagree, 7 = strongly agree) with one statement: I am satisfied with the quality of care I received in the ER. Over the 14-month study period, 148 ED patients participated (note that data collection was paused for 3 months due to COVID-19). Mean patient age was 49.5 years, 58.8% (n = 87) of patients were female, and 39.9% (n = 59) self-identified their race/ethnicity as White non-Hispanic. Of the study sample, 25.0% (n = 37) presented with abdominal pain, 23.0% (n = 34) presented with back pain, 27.7% (n = 41) presented with chest pain, and 23.0% (n = 34) presented with a headache. Our results show that patient perception of diagnostic certainty was a significant predictor of positive affect. Multiple regression analysis (R2 = 0.18, p < 0.001) revealed that perception of diagnostic certainty (β = 0.39, 95% confidence interval [CI] = 0.24 to 0.56) was more strongly associated with positive affect than patient self-reported pain (β = –0.10, 95% CI = –0.26 to 0.07), length of stay (β = 0.02, 95% CI = –0.15 to 0.18), number of tests conducted during stay (β = 0.09, 95% CI = –0.08 to 0.26), number of follow-up actions prescribed (β = 0.07, 95% CI = –0.09 to 0.23), or type of complaint (β = –0.05, 95% CI = –0.21 to 0.11; see Table 1). Patient perception of diagnostic certainty was also significantly associated with patient satisfaction. Multiple regression analysis (R2 = 0.21, p < 0.001) revealed that perception of diagnostic certainty (β = 0.38, 95% CI = 0.22 to 0.54) was more strongly associated with patient satisfaction than patient self-reported pain (β = –0.13, 95% CI = –0.29 to 0.03), length of stay (β = –0.04, 95% CI = –0.20 to 0.13), number of tests conducted during stay (β = 0.19, 95% CI = 0.03 to 0.36), number of follow-up actions prescribed (β = 0.16, 95% CI = 0.00 to 0.32), or type of complaint (β = –0.11, 95% CI = –0.27 to 0.05; see Table 1). Based on 5,000 bootstrapped samples, a mediation analysis indicated that greater diagnostic certainty was significantly correlated with greater positive affect (b = 0.21, 95% CI = 0.13 to 0.30, p < 0.001) and, in turn, greater positive affect was significantly associated with greater patient satisfaction (b = 0.64, 95% CI = 0.48 to 0.79, p < 0.001).9 In other words, perceptions of diagnostic certainty increased satisfaction by increasing patients’ positive affective experience (indirect effect: b = 0.14, 95% CI = 0.07 to 0.23). These findings do not suggest that diagnostic certainty is the only, or the most, important determinant of patient affect or satisfaction; these models do not explain the majority of variance in patient affect or satisfaction. However, these findings do suggest that perceptions of diagnostic certainty play a significant role in a patient's experience and deserve further consideration. By highlighting the importance of a patient's perceived diagnostic certainty on overall satisfaction with care, this study adds to a growing literature investigating the subjective measures of patient experiences that contribute to patient satisfaction with health care. Specifically, when patients feel a greater sense of certainty regarding the diagnosis of their health condition, they also have greater positive affect and report higher satisfaction with care received. Notably, a patient's sense of diagnostic certainty is associated with their satisfaction of care—above and beyond several standard, objective measures, including length of stay, number of tests run, and number of follow-up actions prescribed. While our findings are limited by relatively small sample size and should be viewed as preliminary, we believe that our findings have important implications for ED physicians. ED physicians are trained to assess for and rule out “worst-case” diagnoses and generally not to evaluate conditions that are more appropriate for an outpatient setting. For example, ED physicians focus on “ruling out” a myocardial infarction for a patient presenting with chest pain, rather than “ruling in” any other sources of the pain. Although this approach is standard and often optimal for the ED setting, additional testing, either to rule out worst-case diagnoses or done mainly to reassure patients, may still not provide patients with sufficient diagnostic certainty.10 Rather than performing additional potentially low-value testing, ED physicians may consider simple interventions, such as setting better expectations about the level of diagnostic certainty they believe is possible or changing the ways in which they communicate any degree of diagnostic uncertainty at discharge. Importantly, we do not suggest that this should necessarily change how ED physicians approach the diagnostic evaluation of patients with these or other undifferentiated conditions; overtesting for the sake of certainty could, itself, have significant untoward consequences. However, our findings should factor into how ED physicians communicate the diagnostic approach in the ED and how they frame any remaining uncertainty. Finally, our study raises important questions with respect to the appropriateness or utility of current measures of patient satisfaction in the ED. If such a disconnect exists vis-à-vis the basic goal of an ED evaluation between patient and physician, then evaluating patient satisfaction after an ED visit may be capturing aspects beyond the control of the treatment team. Such a focus on patient satisfaction may then actually be harmful—for example, if physicians were to order additional testing simply to address the concern over patient discomfort with diagnostic uncertainty. These aspects are important when considering the utility of measures of patient satisfaction in the ED setting.