Research

Assessing Efforts to Curb Inappropriate Use of the Emergency Room

Executive Summary

  • Insurers and state Medicaid programs have, over the past several years, implemented a variety of financial incentives to discourage, for potentially non-emergent health concerns, the use of emergency departments (EDs) in favor of other care settings, such as physicians’ offices, urgent care centers, and retail clinics.
  • Recently, some insurers have attempted to increase such deterrence by implementing a policy of retroactive review for ED claims with certain discharge diagnoses and potentially denying coverage for those visits determined to be “inappropriate.”
  • There are concerns that such policies may inadvertently discourage appropriate use of EDs, resulting in individuals not getting needed care in a timely manner. Further, this policy may disproportionately impact minority populations.
  • Several studies indicate that the primary reason for using an ED for potentially non-emergent health concerns is a lack of access to primary care providers. An alternative solution to manage ED usage better, therefore, may be to increase access to alternative care settings, particularly in the evening and on weekends, and to provide transportation assistance to such locations.

Background

For decades, annual visits to an emergency department (ED) have grown faster than our nation’s population, indicating people are visiting EDs more frequently each year. This increased frequency is occurring despite a lack of indication that the number of emergencies is increasing, as shown in this primer on ED trends.

Given that rising health care costs are one of our nation’s biggest challenges, there are numerous reasons for health care stakeholders to be concerned with increased ED use, primarily the much greater cost to treat an individual in an ED compared with the cost at a doctor’s office, urgent care center, or retail clinic. Private insurers in particular are understandably concerned, given that they cover roughly half of all ED expenditures despite privately insured individuals accounting for only 35-40 percent of all ED visits.[1] That said, care in EDs still accounts for just 2 percent of national health expenditures.[2]

As with most health policy challenges, there is no silver bullet to solve the health expenditure crisis; the problem likely requires a multitude of smaller changes. One such change includes decreased use of low-value care—services for which the costs outweigh the benefits they provide. Sometimes care may be considered low-value simply because of the location in which it is provided—a location that, for numerous reasons, may charge higher rates than other sites of care. If the same care were provided at a lower-cost, it may then no longer be considered low-value even though the benefit it provides has not changed. Unfortunately, despite policy experts agreeing on the importance of decreasing the use of low-value care, there are myriad reasons why care might be sought in a particular location, such as the ED, even when the patient knows the cost will likely be higher than if the same care were received elsewhere. Some of those reasons include proximity and the ability to get to one location versus another; as well as the ability to get an appointment, including how soon and whether that appointment is available at a time that is convenient for the patient (for example, without missing work, which can be difficult for some to have approved). Some patients may technically have access to other options, but they do not have other options practically, and incentives or penalties to encourage the other options “available” to them may not remove the barriers they may be facing.

Efforts to Curb Inappropriate Use of Emergency Departments

Most insurers have employed various tools to better manage ED use. Primarily, they have used financial incentives: charging higher cost-sharing for care in EDs than for care in physicians’ offices or urgent care clinics. As of January 1, 2018, as many as 21 state Medicaid programs have also instituted similar financial tools for children and/or adults, including adults who qualify for Medicaid coverage on the basis of being a parent.[3]

Multiple studies, however, have found these policies to be largely ineffective, particularly for lower-income individuals, which may seem counter-intuitive.[4],[5] The primary reason these policies fail to alter people’s behavior amounts to the fact that most people do not use EDs when they have expedient and affordable access to other sources of care and/or when they do not believe their medical concern is an emergency that needs immediate attention. As noted in this AAF primer, research has shown that ED utilization tends to increase during times when physician offices are typically closed. One study found that 75 percent of ED visits among children in 2012 were at night or on a weekend, and that, regardless of insurance status, this lack of access elsewhere (whether real or perceived) was the most common reason given for visiting an ED for non-urgent conditions (i.e. conditions that can wait to be treated for up to 24 hours).[6] In other words, most people go to EDs for care when they feel they have no other option. As a result, there may not truly be enough inappropriate use for this type of financial lever on its own to substantially reduce ED visits.

Denial of Coverage for “Inappropriate” Use of EDs: A Case Study

Possibly because of the ineffectiveness of increased cost-sharing policies, some insurers are now seeking to reduce unnecessary ED use by implementing more stringent policies—namely, by refusing coverage of ED services entirely if the insurer determines the ED use to be inappropriate.

One insurer began implementing this policy in 2015 and is slowly expanding it. The insurer’s justification for such a policy is based on a study that found that between roughly 14 and 27 percent of ED visits could be handled outside of an ED at more appropriate sites of care, such as urgent care centers, walk-in doctors’ offices, and retail clinics.[7] Based on this percentage and the insurer’s internal cost and claims data, the insurer estimated that $1.2 million in costs are incurred each day from unnecessary ED visits; these estimates have not been able to be verified, though.[8] Further, the Centers for Disease Control and Prevention (CDC) classified only 5.5 percent of ED visits as “nonurgent” in its latest report on the subject (though, more than 30 percent had no classification, making it difficult to determine the true percentage of nonurgent visits)[9]. These figures are regularly referenced by both insurers and physicians groups. [10]

Initially, the insurer developed a list of conditions it considers “nonemergent”; if an individual’s final diagnosis matched one of the diagnoses on the list, the insurer would review the claim, including the patient’s presenting symptoms, and if the visit was determined to be “inappropriate”, would deny coverage. More than 12,000 claims (5.8 percent) were denied from July 2017-December 2017, leaving the patient to pay the entirety of the bill.[11] While beneficiaries were warned their ED visit could be denied if it turned out not to be an emergency, they were not provided any information regarding which conditions were on the list, which in one state contained almost 2,000 diagnoses.[12] Some examples of diagnoses that would be considered nonemergent by the insurer included urinary tract infections, acute gastritis, headaches, seasonal allergies, cellulitis, athlete’s foot, chest pain upon breathing, suture removals, and common cold symptoms.[13]

In 2018, the insurer made some changes to its policy to reduce the number of visits that would be reviewed for possible denial. The following circumstances are now all considered by the insurer as “must pay” visits: the patient was directed to an ED by a health care provider; the patient is under age 15; the patient’s home address was more than 15 miles from an urgent care center; the visit occurred during certain weekend hours or on a major holiday; the patient was travelling out of state; the patient received any kind of surgery; the patient received intravenous (IV) fluids or IV medications; the patient received an MRI or CT scan (though, interestingly, a year prior the insurer had notified providers that an MRI test conducted in the ED would be cause for retroactive review because they believed such tests were overused[14]); the visit was billed as urgent care; or the visit was associated with an outpatient or inpatient admission.[15] The insurer stated they would begin requesting a medical record each time before considering whether to deny a potentially nonemergent claim.

The early results of this policy change indicate that most visits are still being covered. A congressional investigation found that 10-20 percent of ED visits were reviewed but only 4-7 percent of claims were denied, with an estimated cost of nearly $3.3 million in claims not paid.[16] A majority of denied claims appealed by patients, however, were ultimately overturned. This investigation also found that the rate of denial dropped dramatically following the adoption of the changes made to the policy in 2018, with zero claims denied in March 2018. Of course, policies like this could still be having a deterrent effect if patients know that nonemergent ER visits could be denied.

Another insurer recently implemented a similar policy of retroactive review for potential denial of coverage. This insurer may review all ED visits for potential retroactive denial, particularly if the member calls the insurer’s 24/7 nurse-line and, against the nurse’s advice, decides to seek care in an ED anyway.[17]

Concerns with Efforts to Curb ED Use

While the examples described above involve diagnosis-based insurer scrutiny of ED visits rather than automatic diagnosis-related denials, research indicates that efforts to reduce potentially inappropriate ED use are likely also to reduce appropriate ED use. And the more punitive the deterrent, the more likely it is to discourage appropriate use along with inappropriate use. Several studies have shown that the symptoms for many conditions ultimately determined nonemergent have tremendous overlap with the symptoms of conditions that do need immediate care. For example, the symptoms of heartburn and a heart attack may feel very similar, and the Mayo Clinic notes that “even experienced doctors can’t always tell the difference from your medical history and physical exam.”[18] Given that every minute is of vital importance during a heart attack, not going straight to an ED when someone thinks they are having a heart attack might be the difference between life and death.

Several studies indicate that ultimate diagnosis is not an appropriate way to measure whether a visit to an ED was appropriate. In 2014, a study found that only 6.3 percent of ED visits were determined, based on discharge diagnosis, to be treatable by a primary care provider; however, the primary complaints reported for those visits were the same as those in 88 percent of all ED visits, more than a quarter of which required immediate or emergent care, hospital admission, or immediate surgery. [19] Another study found 16 percent of ED visits resulted in discharge diagnoses that would be determined nonemergent based on the aforementioned insurer’s list of nonemergent conditions, but the symptoms presented in those cases were again the same as those in 88 percent of ED visits, 65 percent of which received emergency care.[20] Furthermore, it is very likely that in the event an individual does not receive necessary care in a timely manner, the condition may worsen and ultimately cost more to treat, if treatment is even possible later, than it would have cost to treat the individual in an ED at the onset of the condition.

Furthermore, the potential for an “ultimate diagnosis” deterrent policy to disproportionately impact minority and lower income populations is worthy of consideration. These individuals typically use EDs more often and are less likely to have a regular primary care provider. According to the Agency for Healthcare Research and Quality (AHRQ), Black and Hispanic individuals have historically been less likely to have a regular source of primary care compared to Whites.[21] Another study found that the total time burden for receiving ambulatory medical care was up to 28 percent greater for racial and ethnic minorities and unemployed individuals.[22] Additionally, even if an individual has access to a primary care physician, their job (which may not provide sick days or time-off when their doctor’s office is open) and other social factors (lack of transportation options, e.g.) may also impair their ability to access care outside of an ED. These types of obstacles are more typical for lower-income individuals. While these visits may legitimately be nonemergent, it is not clear that retroactive ultimate diagnosis policies effectively encourage patients to seek non-ED care as much as they discourage ED use.

Alternative Solutions

If the primary reason for inappropriate use of EDs is a lack of access to alternative, more appropriate care settings, then the best way to shift care away from EDs is likely by providing greater access to other care settings at all times of the day. Insurer inducements for patients to become familiar with these alternative sites of care and confirm a primary care provider, for example, are now commonplace. These initiatives may require additional care locations or extended hours and/or staffing at existing locations. It may also entail providing transportation options, such as through ride-sharing services, to individuals unable to get to a doctor’s office or urgent care center. A recent collaboration between the Washington state Medicaid program and hospitals and physicians shows that efforts to enhance care coordination, improve access to primary care, and educate patients about appropriate resources for care can successfully reduce ED visits. In the first year of the “ER is for Emergencies” campaign, ED visits declined by nearly 10 percent and the rate of “low-acuity” visits decreased by 14 percent.[23]

Conclusion

Health care costs have been rising for decades, and the trajectory is unsustainable. As such, both private insurers and the government—through Medicare and Medicaid—have long been looking for ways to reverse that trend and reduce costs, primarily by seeking to transition to high-value care and away from low-value care. Part of what makes some care low-value is simply that it is received in a more expensive setting than it needs to be, such as in EDs. Payers have sought to discourage inappropriate use of EDs through a variety of financial incentives and by increasing the availability of alternative means for receiving care. Recent efforts adopted by some insurers, though, may do more harm than good by potentially discouraging even appropriate use of EDs. Providing greater access to alternative care settings may be a more effective way to transition care out of EDs and reduce costs, without the potential for unintended adverse effects.

[1] https://www.americanactionforum.org/research/primer-examining-trends-in-emergency-department-utilization-and-costs/#_edn68 https://www.americanactionforum.org/research/primer-examining-trends-in-emergency-department-utilization-and-costs/

[2] https://www.americanactionforum.org/research/primer-examining-trends-in-emergency-department-utilization-and-costs/#_edn72

[3] https://www.kff.org/medicaid/state-indicator/cost-sharing-requirements-for-selected-medicaid-services-for-section-1931-parents-january/?currentTimeframe=0&sortModel=%7B%22colId%22:%22Location%22,%22sort%22:%22asc%22%7D

[4] https://jamanetwork.com/journals/jamainternalmedicine/fullarticle/2091743,

[5] https://www.healthaffairs.org/doi/10.1377/hlthaff.2009.0906

[6] Renee M. Gindi and Lindsey I. Jones, Reasons for Emergency Room Use Among U.S. Children: National Health Interview Survey, 2012, U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics, NCHS Data Brief: No. 160, Hyattsville, MD, July 2014.

[7] https://www.healthaffairs.org/doi/abs/10.1377/hlthaff.2009.0748

[8] http://file.anthem.com.s3-website-us-east-1.amazonaws.com/04591CAEENABC.pdf

[10] https://www.acep.org/globalassets/new-pdfs/advocacy/anthem-response-to-aha-acr-and-acep-final.pdf

[11] https://www.acep.org/globalassets/new-pdfs/advocacy/07.17.18anthemcoveragedeniedmccaskill.pdf

[12] http://mediad.publicbroadcasting.net/p/kwmu/files/mo_er_member_letter_2017.pdf, https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2707430

[13] https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2707430

[14] https://www11.anthem.com/provider/va/f5/s1/t0/pw_g255375.pdf?refer=ahpmedprovider&state=va (page 14)

[15] https://www.acep.org/globalassets/new-pdfs/advocacy/07.17.18anthemcoveragedeniedmccaskill.pdf

[16] https://www.acep.org/globalassets/new-pdfs/advocacy/07.17.18anthemcoveragedeniedmccaskill.pdf

[17] https://www.optimahealth.com/documents/provider-manuals/plan-management-plan-ref-provider-manual.pdf

[18] https://www.mayoclinic.org/diseases-conditions/heartburn/in-depth/heartburn-gerd/art-20046483

[19] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3711676/

[20] https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2707430

[21] https://archive.ahrq.gov/research/findings/nhqrdr/nhdr11/chap9.html

[22] https://jamanetwork.com/journals/jamainternalmedicine/fullarticle/2451279

[23] http://wsha.wpengine.com/quality-safety/projects/er-is-for-emergencies/

Disclaimer