Inpatient Subspecialty Consultations: A New Target for High-Value Pediatric Hospital Care?

Identifying practice variation across providers is a key first step in identifying low-value practices.¹  Multiple studies of adult inpatients using Medicare administrative data revealed wide variation in consultation use across hospitals that persist after adjustment for patient case-mix and demographics^2,3  and is influenced by hospital size, location, ownership status, and geographic region.³  A single-center study similarly found that the number of consultations per admission in the top quartile of consultation use was nearly sixfold that of the bottom quartile and noted increased consultations on nonteaching hospitalist services (relative to teaching services) and for patients with Medicare and private insurance (relative to patients with Medicaid).⁴ 

In contrast, pediatric literature describing consultation variability is limited. A single-center descriptive study of 200 PHM patients with uncertain diagnoses revealed highly variable consultation practices, with a range of 0 to 8 consult services per patient.⁵  In another study surveying 269 pediatric hospitalists, 27% of respondents reported managing classic, uncomplicated patients with Kawasaki disease independently, whereas the remainder reported typically requesting subspecialty consultation.⁶  Further research is needed to better elucidate current consultative practices, their underlying drivers, and their impact on care.

Consultations are often critical to informing diagnosis, guiding management, and establishing care with outpatient providers. When used optimally, they can also prevent unnecessary resource use. Endocrinology consultation for hospitalized adults with diabetes has revealed improved glycemic control, decreased morbidity, and reduced length of stay.^7–9  Similarly, infectious diseases consultation for Staphylococcus aureus bacteremia in adults has been repeatedly associated with decreasing relapse rates, 30- and 90-day mortality, and 5-year mortality or recurrence.^10–12  Other areas in which the benefit of consultations for hospitalized adults has been demonstrated include palliative care,¹³  behavioral health consultation,¹⁴  dermatology consults for cellulitis,¹⁵  and inpatient heart rhythm services.¹⁶ 

Among studies of hospitalized children, there is limited evidence supporting improved investigation and management of S aureus bacteremia in children with infectious disease consultations,¹⁷  although no differences in 30- or 90-day recurrence or mortality were detected. In a small study of children with enterococcal bacteremia, patients who received infectious disease consultation had lower 1-year mortality.¹⁸  Additionally, a single-center retrospective cohort study revealed that palliative care consultations reduced the likelihood of being in the ICU on a given hospital day,¹⁹  and Bayesian analysis of a single-center randomized clinical trial of a consultation service for children with medical complexity compared with usual care revealed likely decreases in total hospital days, hospitalizations, and health system costs.²⁰  Establishing which consultation practices add value and which should be avoided will optimize patient outcomes and conserve valuable resources.

Although subspecialty consultations clearly benefit patient care in many scenarios, like all medical interventions, they can also lead to unintended consequences. We propose 4 domains in which inpatient subspecialty consultations may detract from health care value.

Some consultations lack clinical benefit and trigger downstream cascades of excess resource use. In a retrospective cohort study of >700 000 admissions to adult hospitalists among Medicare beneficiaries between 2013–2014, patients admitted to high-consulting (top quartile) hospitalists had longer lengths of stay and more costly admissions and were more likely to see outpatient specialists at 90 days after discharge, with no significant differences in 30-day mortality or readmissions, even after adjustment for patient case-mix.²¹  Multiple studies have revealed low adherence to consultant recommendations,^22–24  suggesting opportunities for primary teams to use consultations more judiciously. It is imperative to better delineate the pressures that hospitalists perceive to request consultations beyond the need for clinical guidance (such as medicolegal concerns, deference to subspecialists, and institutional norms) to establish how to optimize the use of this valuable resource. Additionally, it is plausible that well-intended consultants may at times feel compelled to provide testing recommendations that exceed the clinical need in order to satisfy requests of hospitalists and families, leading to overtesting and overdiagnosis.²⁵  A research agenda for consultation value should include exploration of the mechanisms by which consultation use may drive excess resource use.

An association between consultations and delays in care has been explored among hospitalized adults,^26,27  including a retrospective cohort study of patients who underwent surgery for an isolated hip fracture.²⁶  Investigators compared outcomes related to the presence or absence of a preoperative specialty consult and, in multivariable analysis, found that consults were associated with delayed surgery beyond 24 and 48 hours of presentation, extended length of stay, higher 30-day readmission rate, and a nonsignificant trend toward increased 30-day mortality. These findings highlight the importance of both requesting consultations thoughtfully and improving the process of consultation to minimize delays.

Although communication failure is a key contributor to adverse events in health care,²⁸  the role and impact of miscommunication during inpatient subspecialty consultations is understudied, particularly in the pediatric population. Evidence suggests that consults are often missing key information,²⁹  and many providers lack formal training in consult communication practices.³⁰  To date, efforts to improve communication are limited to single-center educational and quality improvement interventions to improve the quality of consults requested by trainees.^29,31,32  Although this is an important start, it pales in comparison with multicenter studies in the pediatric inpatient setting revealing that improved hand-off communication within medical teams can decrease adverse events.^33,34  Because consultant communication similarly involves the exchange of patient-level information, rigorous studies are needed to elucidate the degree to which miscommunication with consultation services may be propagated and develop data-driven interventions that improve patient safety.

In a single-center study among adults with prolonged hospitalizations, researchers found a strong, negative correlation between the number of inpatient consultations and patient satisfaction with physician communication.³⁵  Additionally, in a qualitative study of consultations in the adult ICU setting, two-thirds of families reported having no direct contact with consultants, and families often learned about consultations after they had already occurred.³⁶  Examining this in the pediatric inpatient setting would determine if the American Academy of Pediatrics priorities of shared decision-making and family-centered care^37,38  are being fully realized.

In reconsidering the case of D.M., who presented with fever of unknown origin, did the hospitalist achieve high-value care by consulting subspecialists using an incremental approach, or was her colleague wise to contact all potential consultants after admission? We currently lack sufficient evidence to determine the impact of initiating a consultation on the value of care, making it difficult to answer this question and many others related to consultation practices. The following represent key priorities for research efforts to uncover the value of subspecialty consultations for hospitalized children:

1. Describe interhospital and intrahospital consultation practice variation and investigate the physician, patient, and systems factors that drive them.

2. Identify the best metrics for determining the value of consultations in the pediatric population.

3. Investigate how subspecialty consultations influence patient outcomes and resource use by using robust experimental designs for conditions of interest (for example, osteomyelitis, Kawasaki disease, and bacteremic urinary tract infection).

4. Investigate how subspecialty consultations impact patient and family experience and identify consultation practices that improve patient and family satisfaction.

5. Delineate barriers and facilitators of high-value consultations.

6. Develop and test strategies to enhance existing systems of consultation delivery.

Given the central role of consultations in the practice of hospital medicine and the paucity of pediatric-specific data, this agenda offers a high-impact target for broadly enhancing the value of pediatric inpatient care.