ED Utilization: Addressing the Ongoing Industry Challenge

Ian Chuang, MD Ian Chuang, MD
SVP, Chief Analytics Officer

The appropriate utilization of an emergency department (ED) as a place of health care services remains an ongoing industry challenge. The reasons and circumstances for this persistent issue are varied for consumers, providers and payers. The appropriateness of an ED visit inferred from administrative claims data after the fact is often too categorical in nature, missing the nuances of what a patient was experiencing – including their symptoms or thought process that led them to seek assessment and care at the ED. Additionally, medical benefits can be confusing, and many consumers are not fully aware of the policies and financial terms governing ED utilization and payment coverage.

Primary care physicians have moved away from the “old days” of being the sole 24/7 contact for on-call services. Answering services for physician practices often utilize pre-recorded messages that absolve them of responsibility by instructing callers: “if you feel this is an emergency and require immediate attention, please proceed to the nearest ED.” Ironically, what used to be full-service primary care has now become a buy-up service called Concierge Care, where paying patients get 24/7 access to a physician via phone, text or e-mail.

As a result, payers and self-insured employers continue to see high ED utilization rates, often for non-emergent health care needs, resulting in higher copayment and out-of-pocket cost of ED use. While some consumers are aware of these financial implications prior to visiting the ED, many only learn once they receive an explanation of benefits statement after their visit. Now, payers such as Anthem are retrospectively denying coverage payment for a target list of diagnoses. Anthem predicts to overturn some of these denials when more detailed information regarding ED visit circumstances, type of care received, and reasoning becomes available. However, the responsibility has shifted to the consumer—the least informed individual—to justify their use of ED services.

Options for consumers seeking urgent or after hours care outside of the ED model have expanded with the availability of new facilities and trends in urgent care centers, retail pharmacy care centers, and virtual telehealth services. But again, seeking these newer alternatives requires the consumer to be informed or guided to access these emerging treatment centers instead of the ED.

Can timely engagement with consumers focused on the reasons why they seek care from ED help educate and provide specific directions for ED service access? Can this kind of outreach be provided without negatively impacting those who need and use ED services appropriately?

Given that the use of ED is a combination of signs (what the patient is seeing) and symptoms (how a patient feels) as well as a cognitive decision that the patient needs to seek medical care, and has opted for the ED, can this decision pattern be deduced and used to guide the prioritization of outreach? Can it be leveraged to determine the interventions that are likely to both educate and guide the patient for subsequent or similar circumstances?

Going Deeper: Applying NYU Ambulatory Care Sensitive Condition Logic to Guide Care Management Prioritization and Outreach

The NYU Center for Health and Public Service Research developed an algorithm to help classify ED utilization. This research-based probability profile highlights the most common emergency department (ED) visit diagnoses, putting into context non-emergent vs. emergent visits, and further characterizes emergent visits to avoidable, preventable, or primary care treatable. The ambulatory care sensitive conditions (ACSC) categorization is determined by analyzing the diagnosis in the context of the clinical data surrounding the care. This study also recognizes that for many of the diagnoses, there can be a distribution of clinical scenarios, with some scenarios spanning emergent to non-emergent in context.

Essentially, the NYU study shows that for a given diagnosis, the specific health situation can be considered non-emergent some of the time, and other times emergent. Further, when emergent, the health situation can be considered preventable or avoidable, or in some cases, even primary care treatable. In essence, this study was less black-and-white for many diagnoses when it came to determining a definitive “yes” vs. “no” decision point for an ED visit.

When analyzing and profiling individuals in terms of ED visit categorization, the classifications can be used as thresholds for the desired trade-off between sensitivity and specificity. Our objective is not to make a retrospective coverage determination; the intention is to look for ED usage patterns that can inform and guide care management outreach and interventions. This approach can accept a certain degree of false positive identified individuals. The desire is to identify the patterns where further clinical assessment and root cause analysis can identify and address the drivers that lead to an ED visit for a specific individual. If ED visits are primarily tied to treatable conditions with a primary care provider, and there’s a lack of a PCP relationship or the PCP isn’t accessible, addressing this specific circumstance will help ensure those care needs are addressed before the individual seeks care at an ED.

Where and how does this differ from conventional consumer engagement approaches around their ED usage? While this approach isn’t a replacement to current avenues, it can help ensure that the high touch model of direct consumer engagement can be more effective and relevant to the consumer. Additionally, the traditional mass mailing approach of sending general ED policies and coverage guidelines to individuals may be a lower cost engagement strategy. Targeting a list of frequent ED utilizers without consideration to the specific reasons they sought ED care may also prove to be an effective outreach option.

Would it be valuable to follow-up and engage a consumer at the same time or shortly following an ED visit, so there is immediate relevance and context for engagement? Should all individuals with recent ED visits be contacted? The answer is no. Simply put, this approach is overkill; it is costly and unnecessary for the majority of consumers during their first ED visit in a plan year. However, waiting for an ED overutilization pattern to emerge for a specific member reduces the opportunity to intervene and reduce these repeat visits and avoid the associated, unnecessary costs.

By applying the NYU ACSC approach to the ED visit of concern, the pattern matched to an individual’s past ED utilization profile can help prioritize and inform the possible context for ED usage. This approach improves both efficiency and effectiveness of consumer engagement for ED utilization. It’s more efficient as the outreach and direct engagement of the consumer is only applied to those with a pattern of ED overutilization, suggesting the targeting of addressable circumstances, as opposed to a shotgun approach for all cases. Ultimately, this model can be more effective in understanding the context and pattern of utilization to help identify what can be done to impact specific individuals and direct more appropriate interventions toward identified persons and patterns in real time.

Leave a Reply

Your email address will not be published. Required fields are marked *