a b s t r a c t

Background: ‘Surprise billing’, or the phenomenon of unexpected coverage gaps in which patients receiving out- of-network medical bills after what they thought was in-network care, has been a major focus of policymakers and advocacy groups recently, particularly in the Emergency Department (ED) setting, where patients’ ability to choose a provider is exceedingly limited. The No Surprises Act is the legislative culmination to address “surprise bills,” with the aim of promoting price transparency as a solution for billing irregularities. However, the knowl- edge and perceptions of patients regarding emergency care price transparency, particularly the degree to which ED patients are cost conscious is unknown. Accordingly, we sought to quantify that perception by measuring patients’ direct predictions for the cost of their care.

Methods: We conducted an in-person survey of patients in Emergency Departments (EDs) over an 10-month period at two campuses within a large academic hospital system in southern Connecticut. We surveyed a conve- nience sample of patients at the bedside regarding demographics, care seeking perceptions and their estimates of the total and out-of-pocket costs for their ED care. Survey data was linked to institutional hospital finance datasets including actual charges and payments. We then later obtained the actual costs and billed amounts and compared these to the patients’ estimates using a paired t-test. We also analyzed results according to certain patient demographics.

Results: A total of 600 patients were approached for survey, and data from 455 were available for the final anal- ysis. On average, patients overestimated the cost of their care by $2484 and overestimated out-of-pocket cost by

$144; both of these results met statistical significance (p < .005). Patients were better able to predict both total and out-of-pocket costs if they were: college educated or above; unemployed or retired; aged 65 or older; or had private insurance. Uninsured patients could better predict Total cost but not out-of-pocket costs. One in 4 patients reported considering the cost of care prior to visiting the ED. Only 12 patients reported trying to look up that price before coming.

Conclusions: This study is the first to our knowledge that sought to quantify how patients perceive the cost of acute, unscheduled care in the ED. We found that ED patients generally do not consider the price before going to the ED, and subsequently overestimate the negotiated total costs of acute, unscheduled emergency care as well as their out-of-pocket responsibility for care. Certain demographics are less predictive of this association. Notably, patients with Medicare/Medicaid and those with high school education or below were of the furthest off in predicting the actual cost of care. This lends credence to the established trend of patients’ limited knowl- edge of the total cost of healthcare; moreover, that they overestimate the cost of their care could serve as a barrier to accessing that care particularly in more vulnerable groups. We hope that this finding adds useful information to policymakers in sculpting future legislation around surprise billing.

(C) 2022

1. Introduction

* Corresponding author at: Department of Emergency Medicine, MedStar Georgetown University Hospital, 3800 Reservoir Rd NW, Washington, DC 20007, US.

E-mail address: [email protected] (J.M. Gaylor).

There is a paucity of research about cost-consciousness or price sensitivity of patients seeking emergency care. And despite efforts to in- crease price transparency [1], the emergent nature of unscheduled

https://doi.org/10.1016/j.ajem.2022.08.039

0735-6757/(C) 2022

Table 1

Survey questions.

1. Did you consider the price of this visit before coming to the emergency department? (Yes/No/Not sure)
2. How did you decide to come to the Emergency Department? For example, did you make the decision to come yourself, were you referred, or were you told by family/friend to come?
3. Did you consider (or receive) any alternative care options prior to coming to the emergency department? (Yes/No)
4. Did you consider the price of this visit before coming to the emergency department? (Yes/No/Not sure)
5. Did you try to find out the price of this ED visit prior to arrival? (Yes/No)
6. How much do you think the total bill for your ED visit will be? This is after insurance & negotiations kick in. (Specify dollar amount)
7. How much do you think you will have to pay out-of-pocket in total for this ED visit? (including doctor and hospital fees) (Specify dollar amount)
8. Do you have a co-pay for this ED visit? (Yes/No/Not sure) / How much is your co-pay for this visit? (Specify dollar amount)
9. How much is your annual deductible? (Specify dollar amount/I don’t know)
10. Have you been to an ED before? (Yes/No/Not sure)
11. In prior visits to the ED, do you remember getting: One bill / Different bills / No bills / I don’t remember / Refused to answer
12. If you have never been to an ED before: Do you expect to get one bill for the entire visit, two or more different bills, or are you unsure?
13. What is your highest level of education?
14. Are you currently working and/or going to school?

acute care makes shopping for affordability inherently more difficult. With the increasing penetration of high deductible insurance plans [2], patients are finding themselves more exposed to out-of-pocket costs than ever before-as many as 1 in 6 Emergency Department (ED) visits result in at least one out-of-network bill [3]. Moreover these numbers are rising, even for patients who seek care from in- network facilities [4] (See Tables 1-3).

Accordingly, we sought to examine the patient perspective in the context of their Acute care needs. Our primary research questions centered on identifying the patients’ ability to predict the total costs and out-of-pocket costs for their visits. We accomplished this by asking patients what they predicted the total (negotiated, final) charge to their payor for the visit would be, as well as their predicted (final) out-of- pocket amount, and then compared those results to the actual payments made by the payor and the bills sent to the patients for out-of-pocket costs.

1. Materials & methods

This Institutional Review Board-approved study was conducted in two EDs within a large academic hospital system in southern Connecti- cut between January and October 2019. Patients were surveyed at the bedside and were recruited on a convenience sample basis, with re- searchers conducting surveys at varioUS times throughout the day and week. Patients were eligible if they were 18 years of age or older; could provide informed consent to participate (or accompanied by a

health care representative who was able to provide consent); and were likely to be discharged directly from the ED.

This last criterion was used to identify a certain subpopulation of pa- tients for the specific aims of this study. (These patients were identified by their providers who replied in the affirmative to the question posed by the researcher in the department: Do you think the patient in room xx is likely to be discharged directly from the ED?) First, we aimed to identify and quantify patients’ understanding of unscheduled, acute care, primarily from the ED. Second, the complexity of medical billing increases substantially once a patient is admitted, and teasing the costs of ED care from the hospital course overall would result in less rigorous and less relevant data to the research questions proposed. Fi- nally, patients who are likely to be discharged are, at least on average, presenting with lower acuity complaints, and so theoretically had more time and therefore more agency in choosing to which hospital to present, and therefore at least on average performed a more con- scious decision-making process regarding the cost of that care.

• 1. Surveys

Surveys were conducted at the bedside after informed consent was obtained. If a patient’s primary language was not English, an approved, hospital-based electronic live virtual translator was utilized. The ques- tions were designed to assess in simple language patients’ predictions of total and out-of-pocket costs, as well as demographic information. Surveys were conducted in person by authors JMG and EC by reading from an electronic prompt on a portable, hospital-secured laptop or smartphone.

• 1. Analysis

We then analyzed results by the demographics of those surveyed: education level; Employment status; age group; and insurance status. A separate, paired, two-sided t-test was preformed within each group demographic. For example, a t-test was preformed to compare pre- dicted versus out-of-pocket costs for those aged 65 and older versus those under 65.

1. Results

A total of 600 patients were approached for surveying. Data from 455 surveys were ultimately utilized after excluding patients who either did not or could not consent to the study, were admitted to the hospital, or refused to answer key questions regarding their estimates for the cost of care.

On average, patients overestimated the cost of care by $2484 and overestimated out-of-pocket costs by $144; both of these results met statistical significance (p < .005). Patients were better able to predict both total and out-of-pocket costs if they were: college educated or above; unemployed or retired; aged 65 or older; or had private

Table 2

Difference between patient estimate of cost of total bill versus negotiated payment.

	Patient estimate of total bill	Total charges	Expected (adjusted) payment	Difference	p value
Overall (mean)	$4248.17	7395.88	1764.50	+2483.67	0.0012
College educated or above	4414.15	10,890.06	2607.15	+1807.01	0.1783
High school educated or below	4056.55	3361.94	791.69	+3264.86	<0.0001
Employed and/or student	4025.17	3514.95	1070.53	+2954.64	<0.0001
Unemployed and/or retired	4616.13	13,799.40	2909.55	+1706.58	0.3686
Age 65+	5482.75	24,076.41	5154.53	+328.22	0.9274
Age under 65	3960.67	3339.31	941.31	+3019.36	<0.0001
Private insurance	4934.86	12,780.26	3268.26	+1666.59	0.3539
Medicare and/or Medicaid	3636.53	3958.74	774.43	+2862.10	<0.0001
Uninsured	5783.33	2673.86	817.48	+4965.85	0.2532

Table 3

Difference between patient estimate of out-of-pocket cost versus out-of-pocket responsibility.

	Estimated out-of-pocket cost	Patient Responsibility	Difference	p value
Overall (mean)	$301.27	157.54	+143.73	0.0049
College educated or above	338.51	218.97	+119.54	0.0922
High school educated or below	258.28	86.62	+171.67	0.0204
Employed and/or student	362.74	184.94	+177.80	0.0172
Unemployed and/or retired	199.85	112.33	+87.52	0.1243
Age 65+	182.98	134.57	+48.41	0.5126
Age under 65	328.17	163.92	+164.25	0.0073
Private insurance	379.31	331.25	+48.06	0.5813
Medicare and/or Medicaid	162.43	47.07	+115.35	0.0222
Uninsured	1330.88	0	+1330.88	0.0111

insurance. Uninsured patients could better predict total cost but not out-of-pocket costs. One in 4 (114/455 = 25%) patients reported con- sidering the cost of care prior to visiting the ED. Only 12 (12/455 = 2.6%) patients reported trying to look up that price before coming.

1. Discussion

We sought to quantify patients’ perceptions of the cost of acute, un- scheduled care in the ED. We found that patients significantly overestimated the cost of care by an average of $2484, or 141%; and significantly overestimated their out-of-pocket share by $144, or 91%. Moreover, only a fraction of patients considered costs before going to the ED. Subgroup analysis revealed several trends. Notably, that college-educated, retired, older, or privately insured patients were over- all better at predicting the cost of their care. Moreover, patients with Medicare/Medicaid or those with high school education or below had the greatest overestimation in costs.

Previous research has demonstrated that increased cost of care can serve as a barrier to seeking that care [5,6]. In our study, we quantify the amount of money by which patients overestimate the cost of an acute, unscheduled ED visit. It is our contention that, taken together, these findings support the notion that more vulnerable groups, e.g. the publicly insured and those with less higher education, face the potential for greater burdens when seeking emergency care simply be- cause they view that care as more expensive. Finally, the quantification of the amount of over-estimation of care is a novel finding in our perusal of the current literature on this topic, and can serve as a baseline for future research.

Our study had the following limitations. Notably, our study design cannot capture those who decided against seeking care, i.e. we did not survey potential to-be patients at home. Therefore, we cannot quantify to what extent the perceived increased costs of care limit patients’ seek- ing it. Though patients were selected at random and at different times of the day, week, and year, ultimately ours was a convenience sample of only willing participants; it is not clear how patients who refused partic- ipation would have over or under-estimated cost of care. Lastly, our geographic capture area was restricted to our home institutions.

There are competing financial interests as to how ED billing will be regulated at the Federal and state levels, as evidenced by the vocal input from numerous stakeholders including physician advocacy groups, the insurance lobby, and hospital associations. Perhaps least represented in the shaping of this legislation are those most affected by it: the patients. It is our hope that this data will help amplify the pa- tient voice on this issue, and moreover to inform policymakers by providing quantifiable, primary research about the costs of acute care. Finally, it has been previously demonstrated that there are considerable barriers to the discussion of health care costs between providers and pa- tients [7], and we surmise this research may help understand patients’ perception of cost and subsequent motivators when choosing care.

CRediT authorship contribution statement

James M. Gaylor: Writing - review & editing, Writing - original draft, Visualization, Validation, Resources, Project administration, Meth- odology, Investigation, Data curation, Conceptualization. Edwin Chan: Writing - review & editing, Project administration, Investigation, Data curation. Vivek Parwani: Writing - review & editing, Supervision, Soft- ware, Resources, Project administration, Investigation, Data curation, Conceptualization. Andrew Ulrich: Writing - review & editing, Supervi- sion, Project administration, Conceptualization. Craig Rothenberg: Writing - review & editing, Validation, Software, Formal analysis, Data curation, Conceptualization. Arjun Venkatesh: Writing - review & editing, Supervision, Project administration, Methodology, Investiga- tion, Conceptualization.

Declaration of Competing Interest

The authors have no conflicts of interest to disclose.

Acknowledgement

We acknowledge Jodi Mao for her initial assistance with survey testing.

References

1. Centers for Medicare & Medicaid Services. Fiscal Year (FY). Medicare hospital inpa- tient prospective payment system (IPPS) and long-term acute care hospital (LTCH) prospective payment system final rule (CMS-1694-F). August 2, 2018. Accessed March 2, 2021. https://www.cms.gov/newsroom/fact-sheets/fiscal-year-fy-201 9-medicare-hospital-inpatient-prospective-payment-system-ipps-and-long- term-acute-0; 2019.
2. Palosky C, Ducat S. Average family premiums rose 4% to $21,342 in 2020, Benchmark KFF employer health benefit survey finds. Kaiser Family Foundation; 2020. October 8. https://www.kff.org/health-costs/press-release/average-family-premiums-rose-4-to- 21342-in-2020-benchmark-kff-employer-health-benefit-survey-finds/. [Accessed 2

March 2021].

1. Pollitz K, Rae M, Claxton G, et al. An examination of surprise medical bills and pro- posals to protect consumers from them. Peterson-Kaiser Family Foundation; 2020. February 10. https://www.healthsystemtracker.org/brief/an-examination-of- surprise-medical-bills-and-proposals-to-protect-consumers-from-them-3/. [Accessed 2 March 2021].
2. Sun EC, Mello MM, Moshfegh J, et al. Assessment of out-of-network billing for pri- vately insured patients receiving care in in-network hospitals. JAMA Intern Med. 2019;179(11):1543-50. https://doi.org/10.1001/jamainternmed.2019.3451.
3. Amin K, Claxton G, Ramirez G, et al. How does cost affect access to care? Kaiser Family

Foundation; 2021. January 5. https://www.healthsystemtracker.org/chart-collection/ cost-affect-access-care/#item-uninsured-adults-are-more-likely-to-delay-or-go- without-care-due-to-cost_2017. [Accessed 2 March 2021].

1. Reed M, Fung V, Brand R, et al. Care-seeking behavior in response to emergency de- partment copayments. Med Care. 2005;43(8):810-6. https://doi.org/10.1097/01.mlr. 0000170416.68260.78.
2. Dine CJ, Masi D, Smith CD. Tools to help overcome barriers to cost-of-care conversa- tions. Ann Intern Med. 2019;170(9_Suppl):S36-8. https://doi.org/10.7326/M19- 0778.