LOCAL FRONTIERS: Keeping Frequent Fred out of the ED

Dustin Ballard, MD

I’m certain every veteran emergency physician has had a thought process like this one: “Oh cripes, not Frequent Fred again? Back for chest pain … non-cardiac chest pain? … Could there be something new going on? … Apparently not. … Is it possible that something’s been missed? … No, doesn’t seem like it. … Is there anything new I can offer? … Hmmm—perhaps not. … Soooo, how do I get Fred out the door?”

During my residency training at a county emergency department in Central California, thoughts like this often centered on a frequent flyer I’ll call Vincent. He was a self-declared COPDer who was always short of breath, but never for a discernible reason. His lungs were always clear, and he consistently registered 100% on pulse oximetry. We tried every approach—paper bags, oxygen, inhalers, Valium, steroids, even serial arterial blood gases—but Vincent always came back to the ED, sometimes four or five times a day. Clearly, Vincent had profound needs, but despite our best efforts (including psychiatric and social work consultations), we could not adequately address them.

“Kim” also visited us in the county ED. Many times. She was also short of breath without a discernible reason and was way too young to be in the hospital so often. Her family would drop her off on Friday afternoons for a “vacation.” Hers or theirs, we’d wonder. They wouldn’t answer their phone all weekend and then would show up on Monday to find Kim no better and perhaps worse (from “therapeutic” meddling). After several years of well-intentioned interventions, Kim ended up with a tracheostomy.

I’m quite certain Kim’s tracheostomy was preventable, but how? Looking back, I don’t think ED care could have prevented this unfortunate outcome, because the ED is just not a good place to pursue in-depth, multi-dimensional, long-lasting therapy. It’s too hectic for that. Fluctuations in patient volume, staffing considerations, and the acuity of patients with time-sensitive illnesses combine to make ED care of frequent flyers less than optimal. Yet, frequent flyers continue to place an increasing burden on EDs.

Reliable data on the prevalence of frequent flyers is elusive, primarily because an established definition of frequent flyer does not exist. Nonetheless, a 2010 meta-analysis from the Annals of Emergency Medicine (using a “frequency” definition of four or more visits per year) found that frequent flyers make up 4.5% to 8% of American ED patients and account for 21% to 28% of all ED visits.[1] These figures are attention-grabbing; frequent ED users clearly place a significant burden on our safety net system. But, after a deep breath or two, one wonders whether this is a problem we can fix. And if we were to try, would the effort be worth the cost?

ED frequent flyers have been labeled as “unscrupulous” and “uninsured,” and are accused of “unnecessarily clogging EDs by presenting with primary care complaints better treated elsewhere.” But, as the authors of the Annals meta-analysis note, evidence does not support these stereotypes. Frequent flyers are actually a diverse group of people who are much more likely to be insured than not. Their variety makes one-size-fits-all solutions ineffective. From poorly controlled congestive heart failure, to paroxysmal atrial fibrillation, and from chronic pain to migraine syndrome—there are many reasons why someone might develop an ED habit. Besides, skeptics argue, even if we find a way to “fix” the Frequent Freds and Everyday Eddies, dozens of other frequent flyers will simply take their spots. According to some estimates, up to 75% of ED frequent flyers will be replaced within one year.

Let’s set aside this “regression to the mean” phenomenon for a moment and assume that the problem of frequent flyers is worth addressing. First, however, we need to consider an important and related question: Should EDs be in the business of treating non-emergency conditions? Considering that close to 50% of all ED visits are for non-emergent conditions, this is a valid question indeed. (The 50% estimate is based on both a New York University study that employed exhaustive chart review to associate certain diagnosis codes with non-urgent ED visits and on a Kaiser study using data from millions of ED visits.[2,3])

People come to the ED for all kinds of non-urgent conditions, from toenail fungus to big, black, ugly mole-itis. There are plenty of reasons why someone might choose ED care over other options for non-urgent medical problems–including convenience, lack of insurance, and timely access to specialists and specialized imaging.

As medical professionals, I think we all know that the ED–although expensive and sometimes insufferably slow–can accomplish most work-ups more thoroughly than most other venues. And, quite honestly, non-urgent visits keep some EDs in business. But by the time you reach the point, as we have in California, where 20% of patients leave the ED without being seen, your safety net is getting awfully frayed and in need of repair. So, approaches to providing alternate means for non-emergent care, like urgent care and Rapid Care pathways (as in our San Rafael Kaiser) will help. However, given that frequent flyers take up such a significant chunk of ED time and resources, they are potentially a high-yield population for intervention. Recognizing that managing frequent flyers is not easy, let’s nonetheless explore some options.

Hospital-Based Case Management

What if we used ED visit data to identify frequent flyers and then implemented a multidisciplinary approach to manage them with personalized care plans? This type of preventive care, called “case management,” deploys a team of nurses, social workers and physicians to design and manage outpatient care and social support. With this type of team in place, perhaps Charlie COPD gets his exacerbation picked up earlier and is started on prednisone before he needs to come to the ED (for the seventh time this year).

Different forms of case management are in place across numerous locales and health systems. Within Kaiser’s Marin County facilities, for example, multiple programs exist to provide support, advice and medical management to patients who have been identified as high risk or high utilizers. With Medicare incentives now in place to reduce readmission rates, more such programs are surely on the way, but important questions remain inadequately addressed. Can this type of approach work on a consistent basis? Is it cost-effective?

A recent meta-analysis in the Annals of Emergency Medicine reviewed 11 studies of interventions designed to reduce adult ED frequent flyer utilization.[4] Of these, seven studies were of case management programs, two were randomized controlled trials that compared case-management intervention vs. usual care groups, and two were not case-management based. Results across the studies were mixed, with a consistent decrease in ED visits in the intervention groups, offset by the observation that in one of the randomized trials there was also a significant decrease in ED visits for the control group. (There’s that vexing “regression to the mean” problem again.)

The three studies that included cost-effectiveness analysis suggested that case management saved hospital costs—but only enough to pay for the case management program. The studies did not try to account for non-hospital societal costs, including charity care. This consideration is noted by Maria Raven in her companion editorial: “To be sustainable in the long term, any intervention model must demonstrate the ability to pay for itself in reduced health and social care expenditures, including, but not limited to, those of the ED.”[5] Thus, while clearly promising, further work is needed to evaluate how best to design and study case management for ED frequent flyers.

Community Paramedics

This approach is in its infancy. What if we used pre-hospital providers, such as paramedics, to enroll and deliver pre-hospital case management? Could this supplement existing health system programs? Perhaps. As described in a recent New York Times article, the underlying thesis of the community paramedic is that “emergency workers should not wait around for crises to happen, but rather go out and prevent them.”[6] Thus paramedicine becomes a version of case management provided by paramedics, who have the added benefit of the resources and on-scene expertise of the pre-hospital provider.

Who better to prevent unnecessary transport than the transporters themselves? A homeless outreach program implemented by the San Francisco Fire Department, for example, reduced emergency call volume among the homeless by about 75% in 18 months and saved an estimated $12 million. Unfortunately, the program has been on hiatus since 2009, a victim of federal reimbursement structures that reward pre-hospital transport rather than preventive care. Nonetheless, other municipalities are looking at different angles of the same model.

Alameda County has proposed a pilot program that would offer free primary and preventive care to the public at five county fire stations. Called the Fire Station Health Care Portal, the effort envisions stations staffed by a firefighter paramedic, a care coordinator and a nurse practitioner. The stations would provide non-emergency care, 48-hour ED visit follow-up, and 911/211 phone advice and response. If everything goes well, the program could be up and running by next year, but it may not be sustainable without outside or philanthropic funding.

That funding dilemma captures the current challenge of community paramedicine. Nonetheless, if incentives evolve, we may see a rapid blossoming of paramedicine. Ted Peterson, EMS battalion chief for the Novato Fire Protection District, believes this can and should happen. “The fire service,” he says, “has a long history of both standing ready to respond to emergencies as well as working to prevent them from ever happening. This same philosophy can and should be applied to medical care. The fire service is here 24/7/365 with paramedic firefighters that have the training, equipment, resources and access to help people. It is only logical that they be used in the prevention of medical emergencies. Once the high-risk patients have been identified and protocols established, the community paramedics can ‘check in’ on this population and interject preemptively with treatment or referrals to help patients stay out of the hospital. Not only is this possible, it is the right thing to do for our neighbors.”

Predictive Models

As already mentioned, ED frequent flyers are a diverse group with significant turnover, which makes retrospective-based identification and management problematic. Is it really possible to efficiently “react” when most frequent flyers will resolve their frequency issues on their own? Why not use multi-variable models to predict who will become a frequent flyer?

Some evidence suggests that we can predict frequent flyers. Several studies from the UK have derived and validated algorithms to predict hospital admissions and readmissions. One of those studies found that strong predictors of non-elective admission to UK hospitals included age, male gender, history of previous visits, and the quantity of certain types of prescriptions such as analgesics, antibiotics, diuretics, inhalers, and nitrates.[7] To date, such work has not been extended to the American frequent flyer population, but it is certainly possible and theoretically helpful.

Here’s one possible scenario for a predictive case management model. Electronic medical records run regular reports based on an algorithm or identified risk factors and create a list of at-risk patients. A hospital-based case management team reviews these patients and selects some for intervention. A multi-disciplinary team, including community paramedics, implements the interventions, which might include medication and home safety reviews.

Such a model will surely not eliminate the frequent flyer problem, but maybe it will help soften the burden. And that would be good news not just for Frequent Fred, but also for everyone involved.

References

1. LaCalle E, Rabin E, “Frequent users of emergency departments: The myths, the data, and the policy implications,” Ann Emerg Med, 56:42-48 (2010).

2. Billings J, et al, “Emergency department use: The New York story,” Commonwealth Fund Issue Brief (November 2000).

3. Ballard DW, et al, “Validation of an algorithm for categorizing the severity of hospital emergency department visits,” Med Care, 48:58-63 (2010).

4. Althaus F, et al, “Effectiveness of interventions targeting frequent users of emergency departments: a systematic review,” Ann Emerg Med, 58:41-52 (2011).

5. Raven MC, “What we don’t know may hurt us: Interventions for frequent emergency department users,” Ann Emerg Med, 58:53-55 (2011).

6. Johnson K, “Responding before a call is needed,” New York Times (Sept. 18, 2011).

7. Donnan PT, et al, “Development and validation of a model for predicting emergency admissions over the next year,” Arch Intern Med, 168:1416-22 (2008).

---

Dr. Ballard, an emergency physician at Kaiser San Rafael, writes a medical column for the Marin Independent Journal.

Email: Dballard30@gmail.com