A previously healthy 4-year-old boy presented to an orthopedics clinic with a limp and left knee pain. He reported intermittent pain since falling from his scooter 4 weeks ago and that his symptoms acutely worsened after “tweaking” this knee running 3 days ago. Physical examination revealed a small left knee effusion and limited ability to bear weight, but normal range of motion without erythema, warmth, or fever. Plain radiographs and white blood cell count (5.9 × 109/L) were normal, with an undetectable C-reactive protein (CRP) (<0.4 mg/dL). However, his erythrocyte sedimentation rate (ESR) was mildly elevated to 20 (normal, 0-10 mm/h), prompting referral to the emergency department, where knee ultrasound showed soft-tissue swelling consistent with recent trauma and no obvious effusion. Joint aspiration under sedation was performed and was reassuring against septic arthritis with a leukocyte count of 825/µL. However, he was started on broad-spectrum antibiotics and admitted to the hospital for persistent concern for osteomyelitis given the ESR. Magnetic resonance imaging under sedation the next morning demonstrated soft-tissue swelling and muscle edema consistent with trauma or myositis rather than infection. The antibiotics were discontinued. Because of microcytic anemia (hemoglobin, 10.8 g/dL; mean corpuscular volume, 61.9 fL) on the initial complete blood count, a diet history was obtained that revealed the patient to be a “picky eater.” Subsequent laboratory tests revealed several nutritional deficiencies including low iron and undetectable vitamin C and D levels. Literature review revealed that vitamin C and D deficiency can mimic musculoskeletal (MSK) infections.1–3 The patient was started on a daily multivitamin and discharged from the hospital with close pediatrician follow-up. His pain resolved within 1 week and mobility improved, although he remains traumatized from the hospitalization and terrified of doctors’ appointments.
This case highlights some of the specific challenges when faced with diagnostic uncertainty, and how premature closure and communication breakdowns can result in patient harm and low-value care, in this instance multiple blood draws, an invasive procedure, repeated sedation, and unnecessary antibiotics.
Cognitive biases and missed opportunities
Ambiguity arises because of the perception of missing information, and individuals will tend to select the least ambiguous option when faced with a choice, regardless of whether that choice is most likely to bring about the desired outcome.4 In medicine, clinicians will often resolve clinical uncertainty by pursuing diagnoses or treatment plans they are most familiar with and perceive as less ambiguous, even if the likelihood of the working diagnosis or success with treatment is lower than the alternative.5 Termed the “ambiguity effect,” this cognitive bias leads to errors when integrating patient information into management plans because we tend weigh “objective” data like laboratory values over “subjective” data from the patient’s history given their perceived certainty. It also results in premature closure through our reliance on illness scripts and other heuristic driven decision-making processes that seek to impose structure and certainty onto patient presentations.6
In this patient’s case, multiple providers anchored their diagnostic reasoning to the ESR and continued pursuing additional interventions to explain this abnormality. In doing so, they discounted data less consistent with infection as a source for his pain, including the history and ultrasound suggestive of repeated mechanical trauma, his lack of warmth or fever, his normal range of motion, and his undetectable CRP. This occurred despite evidence that CRP is superior to ESR for ruling out MSK infection,7 and despite most decision aids, including our own institution’s, recommending a cutoff of >40 mm/h rather than the laboratory’s upper limit of 10 mm/h.8,9 However, seeing the result flagged as abnormal in the electronic medical record recalled the familiar illness script for MSK infection (ie, a limping child with elevated inflammatory markers), resulting in the misapplication of our institution’s pathway because it provided clear management plan as opposed to the ambiguity of a limping child with a single mildly elevated inflammatory marker and no readily apparent unifying diagnosis.
As it happened, the patient experienced a cascade of low-value and unnecessary interventions culminating in a series of negative studies and a diagnostic dead end. It was only at this point that his care team reconsidered their working diagnosis and “rediscovered” his microcytic anemia, which ultimately led to the correct diagnosis. Although less common than MSK infection, arthralgia and myositis resulting from vitamin C and D deficiencies are well described in the literature and frequently lack the hallmark signs typically associated with scurvy such as perifollicular hemorrhage and gingival bleeding.1–3 Had the team considered the incongruent and ill-defined features of his presentation from the beginning rather than shoehorning his case into a familiar but poorly supported diagnosis, a more thorough history and targeted nutritional laboratories would have resulted in a beneficial cascade of high-value care instead.
Structural drivers of low-value care
Patient outcomes are not simply the result of individual behaviors, but rather the interplay of interpersonal, institutional, community, and policy factors.10,11 This is particularly true under conditions of uncertainty, when clinicians often turn to external supports ranging from colleagues to structured approaches to mitigate this uncertainty and develop a roadmap from which to proceed. Clinical pathways are 1 example of external supports that offer standardized approaches for providing evidence-based care and have been associated with improved patient outcomes and lower costs.12,13 Intended to minimize cognitive biases, they may instead reinforce them if applied incorrectly. As salient and easily accessible tools that simplify the ambiguities of clinical decision-making, patients with uncertain or working diagnoses are often wrongly placed into a specific pathway and treated accordingly. It can then be challenging to identify the off-ramp and consider alternative diagnoses for a patient who is “on the pathway.” For this patient, an overreliance on our institution’s MSK infection pathway further anchored the team to this diagnosis and sustained the subsequent diagnostic momentum and cascade of low-value care.9 Errors in clinical reasoning contributed to its initial misapplication; however, features of the pathway itself contributed, too. Whenever possible, clinical pathways should target patient presentations rather than specific diagnoses (eg “limping child” vs. “MSK infection”) to avoid reinforcing clinicians’ bias toward premature closure. When greater specificity is required, pathways should be developed with clear, easily interpreted criteria and include explicit branch points for reassessment or exiting the pathway.
Shortcomings in communication between the care teams further contributed to this case. After the initial blood work and ultrasound, neither the emergency physician, consulting orthopedic surgeon, nor hospitalist contacted for admission were confident in the working diagnosis of MSK infection. However, this independent yet shared diagnostic uncertainty was never fully voiced nor communicated directly among the care team. Absent a clear alternative explanation, management for a presumed infection continued unabated with the patient undergoing a sedated joint aspiration followed by admission for sedated magnetic resonance imaging. Diagnostic uncertainty is unavoidable in medicine, yet shared acknowledgment of it among members of the care team occurs infrequently and may represent a barrier to quality care.14 The discomfort with the working diagnosis should have been named and addressed more directly by the patient’s care team, including shared decision-making with the patient’s family, particularly before pursuing multiple invasive and costly procedures. Standard workflows should facilitate direct communication between specialists, whereas institutions must ensure a culture of psychological safety so as to better elicit and address diagnostic uncertainty.
Developing the expertise to recognize and avoid cognitive biases is typically characterized as a skill mastered through experience,15,16 with techniques such as diagnostic time-outs and reflective practice frequently suggested as tools to improve care.17,18 However, personal habits and attitudes are insufficient unless bolstered by cultural norms and structural supports, which, as demonstrated by this case, may help to reinforce or mitigate diagnostic errors. This recognizes that the daytime hospitalist’s diligence and clinical acumen was necessary but not sufficient, and that external factors helped her reach the correct diagnosis. For example, the opportunity to observe patients over time, return to obtain further history, gather input from multiple specialists before finalizing her plan, and reframe preliminary diagnoses based on this information is a structural advantage in hospital medicine that is not always available during 15 minute referral clinic visits or busy emergency departments. Developing systems to identify, manage, and address uncertainty in clinical decision-making is the responsibility of both individuals and institutions. This is essential to the critical goals of avoiding premature closure and providing high-value care for patients and families.
Acknowledgments
The patient’s guardian provided written consent to share the case description for the purpose of this published report.
Dr Gehle conceptualized and designed the manuscript, drafted the initial manuscript, and revised the manuscript. Dr Zwemer critically reviewed and revised the manuscript for important intellectual content. Dr Harrison conceptualized and designed the manuscript and critically reviewed and revised the manuscript for important intellectual content. All authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.
FUNDING: No external funding.
CONFLICT OF INTEREST DISCLOSURES: The authors have no conflicts of interest relevant to this article to disclose. Dr Harrison is funded by the Health Resources and Services Administration of the United States Department of Health and Human Services as part of a National Research Service Award (T32HP14001). The funding agency did not participate in this work. The contents are those of the author(s) and do not necessarily represent the official views of, nor an endorsement by, the Health Resources and Services Administration, United States Department of Health and Human Services, or the United States Government.
Comments