Published on

Click Here to download the PDF

In the macro and “dot phrase” era, there are many refrains that appear in the electronic medical record (EMR) with such regularity that we don’t even notice them anymore. Statements like All questions were answered prior to discharge, The patient verbalizes understanding and is comfortable with the plan, and Symptomatic care and over-the-counter treatments discussed are so commonly tacked onto charts that our eyes have been trained to gloss over them. When was the last time one of these statements rose to a meaningful level of consciousness for you?

Such comments do not add much (if any) value for communicating the course of care or our thought processes with other healthcare personnel. Rather, they’re inserted because, like a seatbelt on an airplane, they offer a modicum of comfort and (mostly) an illusion of protection. We sleep a little easier knowing that such phrases reside at the end or our notes because we believe that, should we ever have the misfortune to learn our chart is under the scrutiny of a plaintiff’s attorney, they’ll keep us safe from litigation.

Other than contributing to “note bloat”—a significant, but largely unavoidable nuisance of modern medicine—these overly general, protective statements are mostly harmless. However, when templated, generic instructions infiltrate our patients’ follow-up plan and aftercare instructions, their effects can become decidedly more pernicious.

The most common example of this lies in the instructions and timeline recommended for ongoing care after the patient is discharged from clinic. “Follow-up with your primary care provider (PCP) in 2-3 days,” is the mantra I’ve seen appear with the greatest frequency. This recommendation may soothe us because it’s concise, expedient to include (it’s usually just part of a template), and theoretically protective. The issue is that it’s rarely practical within the confines of the current U.S. healthcare landscape and, even if achievable, it would almost always be bad medical advice. What’s most unfortunate, though, is that many patients actually still trust us and, therefore, take this recommendation seriously.

On initial appraisal, this statement may seem perfectly appropriate. After all, it does cover the important aspects of a good follow-up plan in that it is both time-specific and action-specific. “Follow-up with your PCP in 2-3 days.” It tells the patient who to see and when. This makes the advice more actionable than the still-oft used “Follow-up with your PCP” or highly enigmatic “Return if worse.” The vagueness of these statements renders them nonspecific to the point of meaninglessness. But, the perniciousness of directing patients precisely towards a 2–3-day primary care revisit lies mostly in the fact that it is just specific enough to be taken seriously, while simultaneously being highly impractical and medically inappropriate most of the time.

Let’s begin with the impracticality aspect because it’s most obvious. In a bygone era, before smartphones and DVRs, most Americans had a primary care doctor. They knew their doctor and their doctor knew them. Their doctor could be reached by phone (and even appreciated the call) when their patient came to urgent care in the evening hours. If their patient came in over the weekend, the PCP would make time to see them on Monday morning. Things are different now. The most recent study addressing the topic found that 25% of U.S. adults did not even have a PCP.1 However, this paper examined data from 2015, and personal experience from anyone providing episodic care in the post-COVID age would suggest that this figure is almost certainly considerably higher. Furthermore, this study didn’t delve into the quality of patient-PCP relationships, and I’ve certainly found that the proportion of patients with nominal PCPs whom they’ve never met seems to be growing rapidly and continuously.

In recent years, largely due to secondary effects of the pandemic, the rates of healthcare providers leaving medicine has risen sharply. In 2020-21 alone, 30% of U.S. medical personnel left their positions.2 In fact, 117,000 doctors—nearly 15% of the U.S. physician workforce—left their jobs in 2021.3 An additional 20% of American healthcare workers, according to a survey published in the Mayo Clinic Proceedings in 2021, stated that they intended to leave their current employer.4 This is the so-called “Great Resignation” and, if you haven’t been part of it, you’ve undoubtedly felt its effects—and so have our patients. I see multiple patients each shift presenting with requests for refills of long-term medications for conditions like high blood pressure and hypothyroidism because their PCP has retired or moved on. The waitlist to see a new PCP in my healthcare system is currently longer than 6 months. I’d wager that your experience is similar. And with 45% of physicians aged 55 and older, this problem is unlikely to improve anytime soon.5

Furthermore, this turnover creates illusory PCP-patient relationships even for the patients who have one. I hear story after story of patients waiting weeks or months to be seen at their primary care clinic only to be greeted by a new or cross-covering provider who they’ve never met. In essence, the foundation of primary care—continuity in clinician-patient relationship—has largely gone the way of the 8-track and cassette tape. So, not only is it laughable to imagine currently that most patients can be seen by a PCP they know within 2-3 days, but asking patients to even try suggests we are oblivious to the current crisis affecting the U.S. healthcare system. Proposing that a patient attempt this Herculean task simply furthers frustrations and disillusionment with the entire medical establishment.

More importantly, with the knowledge of the accessibility crisis, it’s worth being honest with ourselves about how many (or rather, how few) of our patients actually need follow-up within this time frame, if at all.

When pondering this question, it’s worthwhile to consider why we have patients with acute issues follow up in the first place. As a guiding principle, the timeline for rechecks for acute problems should be guided by the natural history of whatever concerning conditions remain in our differential diagnosis at the time of discharge. This should be coupled with consideration for the relative likelihood of these disease entities (which hopefully is low if we are discharging the patient), as well as the consequences of these diagnoses, if missed, for the patient’s morbidity and mortality.

To unpack this, let’s consider a few everyday examples.

When we evaluate a child with vomiting, we (hopefully) realize that, while it’s almost always gastroenteritis, a small fraction of these children may be vomiting as an early manifestation of appendicitis. The natural history of gastroenteritis, however, is spontaneous recovery over several days. For the cases in which this pattern of recovery unfolds as expected, why would we compel an exhausted parent to take more time off work and take their now recovered child out of school again to see a pediatrician? Keep in mind the pediatrician is likely struggling to find time to keep up with their essential role as a PCP for their patient panel. The parent, child, pediatrician, and the pediatrician’s other patients are all adversely affected if this recommendation is followed. Conversely, the natural history of appendicitis involves a relatively rapid progression towards rupture, which then proceeds to peritonitis, sepsis, and death. Beginning at 36 hours after symptom onset, the risk of rupture increases by around 5% every 12 hours.6 So, if we are telling parents to get rechecked in 3 days and it turns out that their child has appendicitis, we are telling them to wait until it may be too late. In neither of these hypothetical cases does mandatory follow-up in 2-3 days serve the patient or the subsequent clinician.

Another common example where this advice is problematic can be seen in the follow-up recommendations for most simple orthopedic injuries, such as knee and ankle sprains. These tend to improve over weeks-to-months in most patients without treatment. Occasionally, however, they don’t improve, and patients do require advanced imaging, physical therapy, or even surgery. However, if a patient shows up at their PCP’s office 48 hours after being seen in UC for a knee sprain, not much will have changed and it will still be far too early in the natural history of recovery to determine if the patient is going to need specialist attention or an MRI. If such a visit does miraculously occur, several things tend to happen, and none of them represent high-value or high-quality care.

First, the patient is likely to expect something more to be done (eg, orthopedics referral, prescription analgesics, etc.). Secondly, the PCP is likely to believe that the patient is expecting something more to be done (even if they’re not). If either of these conditions is true, then unnecessary and potentially harmful testing or treatment is a likely result. After all, it’s hard to improve on quality or value by doing more for patients with self-limited conditions.

Perhaps the most common, and consequently frustrating, example of unnecessary short-term follow-up concerns the case of a lingering upper respiratory infection (URI). It probably won’t be hard for you to remember a time when this happened to you: a patient who was seen by a colleague for 4-5 days of cold symptoms returns 2 days later because they’re “still coughing.” This can occur simply because the patient has unrealistic expectations or didn’t get the antibiotic they thought they needed at the first visit. However, more often, patients return because the provider who saw them initially recommended a recheck in several days and the frustrated, still-ill patient simply followed their advice. The natural history for URIs, which we are all hopefully intimately familiar with, involves at least 2 to 3 weeks of cough in many cases.7 Asking a patient to return for re-evaluation before the time of expected natural resolution does nothing but increase the collective frustration of all parties and further crowd clinics with cases of contagious disease.

Now, I’m not dismissing the value of follow-up care entirely; it certainly is appropriate and necessary in the right context. When considering chronic disease care (think diabetes and blood pressure management), regular follow-up with some specific cadence is critical. This is because these are long-term, if not lifelong, conditions and patients usually are treated with an indefinite regimen of daily medications. For example, patients with refractory hypertension and diabetes who are being treated with an ACE-inhibitor and insulin should be regularly reviewing blood glucose and blood pressure logs with their provider and having labs like renal function and hemoglobin A1c checked on a scheduled basis to evaluate the effectiveness and tolerability of treatment.

Compare this with the acute, episodic care of mostly self-limited conditions that we deliver in UC. Patients usually present to UC because of a minor injury or new symptoms. In doing so, they’ve demonstrated that they’re comfortable seeking out care if they have health concerns. Therefore, since we know they’re reliable, it makes the most sense to simply ask them to return or go to the emergency department if things don’t proceed according to the expected natural history of the condition we’ve diagnosed. As long as our instructions are time-specific and action-specific (eg, return here or go to the ED immediately) and we communicate diagnostic uncertainty and the possibility of things not going as planned, these statements offer the most practical guidance.

As a general rule, a good framework for provisional follow-up instructions should take the form of “if/then” statements. For a URI, this may be something like, “I believe you have a viral URI. Recovery commonly takes up to 3 weeks. If your symptoms persist longer than this, return to UC or see your PCP for further evaluation. If you develop shortness of breath, fevers >101°F, pass out, or have other new or significantly worsening symptoms, then seek care immediately in the emergency department.” Adding “If you think you’re having an emergency then call 911” doesn’t hurt either and, while trite, is much better advice than telling a patient with a back strain or sore throat to see their PCP in 2-3 days.

As a final note, it’s worth mentioning that there are certainly some cases where short-term, mandatory follow-up is highly advisable, if not obligatory. Occasionally, it’s because patients need specific procedures like casting or suture removal. There are also higher-risk conditions, like chest pain and serious hand injuries, where close/rapid specialist follow-up is protective for both the patient and ourselves. This group, however, is a small minority of the patients we see.

In most situations, we are sparing patients, PCPs, and our colleagues the stress and risks of likely unnecessary care by foregoing mandatory, short-term follow-up recommendations for most acute issues we see in UC. This is especially true in the wake of the pandemic and consequent healthcare access crisis. In fact, putting this advice in the EMR and communicating it to patients furthers frustrations and the other issues perpetuating the crisis. Our patients have proven that they know how to access care if they feel the need by virtue of simply showing up in our UC centers in the first place. So when we discharge them, let’s give them practical and personalized guidance about where and when they should be seen next. This fosters trust and appreciation, which are far more protective than whatever impersonal and generic follow-up recommendations we might be tempted to plaster on the bottom of our discharge instructions.


  1. Levine DM, Linder JA, Landon BE. Characteristics of Americans with primary care and changes over time, 2002-2015. JAMA Intern Med. 2020;180(3):463-466.
  2. Galvin G. Nearly 1 in 5 health care workers have quit their jobs during the pandemic. Morning Consult. October 4, 2021. Available at: Accessed March 28, 2023.
  3. Definitive Healthcare. Addressing the healthcare staffing shortage. Available at: Accessed April 5, 2023.
  4. Sinsky CA, Brown RL, Stillman MJ, Linzer M. COVID-related stress and work intentions in a sample of us health care workers. Mayo Clin Proc Innov Qual Outcomes. 2021;5(6):1165-1173.
  5. American Medical Association. AMA Physician Masterfile (December 2019).Available at: Accessed March 28, 2023.
  6. Bickell NA, Aufses AH Jr, Rojas M, Bodian C. How time affects the risk of rupture in appendicitis. J Am Coll Surg. 2006;202(3):401-406.
  7. Paalanne N, Honkila M, Mattila S, et al. Duration of clinical symptoms in children with acute respiratory infection. Acta Paediatr. 2023;112(4):813-819.
It’s Time to Think Differently about Follow-up
Joshua Russell, MD

Joshua Russell, MD, MSc, FAAEM, FACEP

Quality and Provider Education, Legacy - GoHealth Urgent Care, is affiliated with the University of Chicago Medical Center in Vancouver, WA, and is Editor-in-Chief of JUCM.