Abstract

Dear Editor:
In January 2016, the Centers for Medicare and Medicaid Services launched two procedure codes that allow clinicians to bill for advance care planning (ACP). 1 In this study, we describe patterns of ACP code use by clinicians among their privately insured patients in 2016.
We performed a retrospective cohort analysis using the Optum© Clinformatics® Data Mart, which contains commercial and Medicare Advantage claims for 13.7 million members of a health insurer in the United States. We included patients with medical coverage in 2015 and 2016, and we identified clinicians billing for ACP based on claims containing code 99497 or 99498.
We found that among 991,064 clinicians, only 7139 (0.7%) submitted an ACP claim in 2016. The median number of ACP encounters per clinician billing for ACP was 2 (interquartile range 1–6). This may be due to limited opportunities for ACP among their patients, or inconsistent use of a novel billing code that has not yet been integrated into their routine practice.
Although primary care clinicians submitted 69.0% of all ACP claims, code use was highest among palliative and geriatric medicine physicians as a proportion of total clinicians within that specialty (12.4% and 8.8%, respectively) (Table 1). Although 82.5% of all ACP claims were submitted in outpatient clinics, palliative care clinicians submitted more ACP claims in inpatient hospital rather than ambulatory settings (72.7% vs. 6.8%, respectively).
Types of Clinicians Using the Advance Care Planning Billing Codes in 2016
ACP claim use by clinician type = number of ACP claims submitted by clinicians of a given type/total number of ACP claims.
ACP claim use within clinician type = number of clinicians of a given type submitting ACP claims/total number of clinicians of a given type.
ACP, advance care planning.
Code use was infrequent even among specialists such as oncologists and nephrologists who routinely care for seriously ill patients. At our institution, the average reimbursement for this service is <$100, which may be an insufficient incentive for physicians who are able to bill for infusions, dialysis, and other high-revenue services. Indeed, adoption was highest among palliative care physicians, who similarly care for many seriously ill patients, but who typically have fewer opportunities to use more remunerative billing codes. Alternatively, early knowledge of the codes may have been more widespread among palliative care clinicians who regularly engage in ACP.
Although palliative care physicians used the codes more frequently than others, their absolute rate of use was low. It is likely that most palliative care physicians are already engaging in ACP; however, nearly 90% of them did not bill for ACP in this patient population. This is a missed opportunity to bill for services provided, although the financial impact of these codes is uncertain in inpatient settings where palliative care clinicians tend to practice.
This analysis has important limitations. First, we could not include Medicare fee-for-service beneficiaries; however, similar trends have been reported in that population.2,3 Second, we could not investigate the reasons underlying limited adoption of the codes. Future qualitative studies should seek to understand clinicians' perspectives on use of the codes, particularly palliative care clinicians who are already routinely engaging in ACP. Despite these limitations, we hope that our findings will stimulate discussion in the palliative care community about the merits of reimbursement for this important service.
