Understanding The Use Of Medicare Procedure Codes For Advance Care Planning: A National Qualitative Study

Abstract

In 2016 Medicare introduced advance care planning Current Procedural Terminology (CPT) codes to reimburse clinicians for time spent providing the service. Despite recent increases, use of these codes remains low for reasons incompletely captured by quantitative research. To further identify barriers and facilitators to code use for Medicare fee-for-service enrollees, we conducted case studies at eleven health systems, including 272 interviews with clinicians, administrators, and key leadership. Five themes related to use of the new codes emerged: code-based constraints to billing, burdening patients with unexpected charges, ethical concerns with billing for discussion of advance care plans, incentives to signal the importance of their use in billing, and increasing both workflow burden and the need for institutional supports and training. Respondents also observed that use was facilitated by health systems’ investment in clinician training and in processes to audit the codes’ use. Our findings suggest that increased reimbursement, strong institutional commitment and support, and streamlined workflow could improve the use of the new CPT codes to document receipt of and ensure access to Medicare advance care planning.

Advance care planning (ACP) is integral to patient-centered care. A systematic review found that as of 2016 only one-third of patients in the US reported completing an advance directive or living will.¹ Authorization of payment for ACP for Medicare fee-for-service providers represents the Centers for Medicaid and Medicare Services’ (CMS’s) recognition of ACP’s value for improving goal-concordant care for patients with serious illness or those facing medical emergencies.² The advantages expected to accrue from proper use of these billing codes include more accurately estimating the prevalence of advance care planning and its impact on the quality of care by better capturing ACP services—a current challenge for payers, policy makers, and researchers, given the heterogeneity of ACP billing and documentation. Moreover, ACP codes compensate clinicians directly for delivering complex ACP and align billing codes with services provided.³ In combination with other interventions, these codes create a standardized way to bill for ACP and monitor its receipt—one that can be used as a quality metric and that creates opportunities to examine patterns in delivery.

To date, quantitative studies have found minimal, albeit growing, use of ACP billing codes for reasons that are incompletely understood.^4–6 ACP code use increased 30 percent from 2018 to 2019—the fastest of any evaluation and management code in Medicare’s fee schedule during that period (from $115 million reported to $150 million reported).⁷ Yet gaps remain in understanding how and why clinicians decide to use ACP billing codes as opposed to using alternative codes or not billing specifically for the service, as well as the role of health systems in framing those decisions.⁸

Most prior studies rely on claims data to characterize clinician factors associated with ACP billing,^9–12 leaving unaddressed the more complex situational, cognitive, and emotional factors that contribute to variation in clinicians’ billing behavior.¹³ Given that integrating ACP codes requires health system investments (for example, training clinicians in service provision, documentation, updating workflow, augmenting electronic health systems, and enhancing auditing), qualitative research is needed to identify clinician- and system-level factors that promote or impede ACP billing. This qualitative study addressed gaps in the literature by identifying key barriers and facilitators to ACP billing as reported by a large, national sample of representatives from eleven health systems.

Background

In 2016 CMS introduced two Current Procedural Terminology (CPT) codes that allow physicians and advance practice professionals to bill for ACP for fee-for-service Medicare across clinical settings. The codes reimburse at a rate of approximately $80–$86 for the first thirty minutes and $75 for each thirty minutes thereafter.¹⁴ Both the relative value unit and reimbursement values for ACP services are comparable to those for commonly used evaluation and management codes in the outpatient setting. Most commercial health plans, including Medicare Advantage, follow the Medicare Physician Fee Schedule and reimburse ACP codes similarly to fee-for-service Medicare, although they are not required to do so. To use these codes, clinicians must dedicate at least sixteen minutes of a face-to-face visit to ACP, which could occur as a stand-alone visit or with other services. Cost sharing (patient copayment) applies to ACP services but is waived if ACP is provided as part of a Medicare annual wellness visit. Eligible services include discussion of values and care preferences and completion of advance directives, although completing forms is not required. However, it should be noted that ACP billing does not capture all service delivery. Some practitioners, such as registered nurses, social workers, and chaplains, may provide ACP but lack the authority to bill for it using the ACP codes. Other clinicians who are eligible to bill with the ACP codes may choose to either bill using codes for increased visit time or complexity or opt not to bill for the service.

Study Data And Methods

This qualitative study used a case-study approach.^15,16 An advisory panel consisting of leaders in palliative care and health services research identified health systems with evidence of prioritizing advance care planning (see online appendix exhibit A).¹⁷ Additional sites were identified from key-informant interviews and publicly available data (appendix exhibit B).¹⁷ To capture the spectrum of implementation barriers and facilitators,¹ health systems were purposively sampled on the basis of ACP prioritization, size, geography, type (for example, academic or public), religious affiliation, and approach to ACP delivery (for example, palliative care–led, integrated, or chaplain-led). We included federally qualified health centers and safety-net hospitals, as these systems may develop different strategies for ACP (appendix exhibit C).¹⁷

DATA COLLECTION

The research team completed qualitative semistructured face-to-face interviews during site visits.¹⁸ At each site we identified a “site champion,” typically the chief medical officer or senior palliative care physician, who was familiar with ACP efforts. Site champions compiled a list of potential participants’ email addresses and purposively sampled them on the basis of role (for example, executive leadership, billing specialist, physician, or nurse), care level (for example, primary or tertiary), and experience with ACP. We included participants with knowledge of ACP integration and billing efforts, including management and billing and coding specialists. Institutional billing guidance, electronic health records (EHRs), and training documents were examined.

The research team recruited participants by email with assistance from site champions. Between August 2018 and December 2019 the qualitative team (authors) conducted face-to-face interviews at participants’ workplaces (or by phone when preferred) and took field notes upon receipt of participants’ oral consent. All interviews were audiorecorded and professionally transcribed. Participants also completed a demographic survey after the interview (appendix exhibit D).¹⁷ This study was approved by the Partners HealthCare Institutional Review Board.

Interviews followed semistructured interview guidelines developed by the research team based on literature review and clinical experience. Open-ended questions probed billing decisions, use of codes, and barriers and facilitators to ACP billing (appendix exhibit E).¹⁷ Additional questions provided data about organizational characteristics (leadership, training, and incentives) and how health systems integrated ACP.

ANALYSIS

Transcripts were uploaded to NVivo 11. Interviews took place in parallel with analysis and continued until thematic saturation was confirmed through deliberation.¹⁶ We created a preliminary codebook deductively based on the interview guide and iteratively revised it for emergent codes inductively.¹⁹ We double-coded 15 percent of transcripts. The codebook was finalized and applied to the remaining transcripts after refinement and deliberation, with Keren Ladin resolving disagreements. Codes were iteratively organized into themes to capture the range of narratives and counternarratives. Findings were shared and confirmed with site champions. Reporting followed consolidated criteria for reporting qualitative research guidelines.²⁰

LIMITATIONS

As with all qualitative studies, ours produced findings that might not be generalizable to systems other than those sampled. However, we purposively selected health systems that were diverse in their emphasis and commitment to ACP to learn more about barriers and facilitators to ACP code use across a large spectrum of systems.

A second limitation is possible response bias, which is a concern with all self-reported data. To address this, interviewers used neutral questions and responses and probed for details. To minimize recall bias, we asked participants about current, not historical, practices.When possible, we confirmed reported billing practices by collecting billing data and requesting demonstrations of billing in local EHRs. Our findings apply to billing practices for fee-for-service Medicare enrollees and might not be generalizable to Medicare Advantage enrollees in plans that did not adopt ACP reimbursement. Finally, variability in billing for ACP is not a proxy for delivery of ACP services.

Study Results

The sample included eleven diverse health systems across the US (exhibit 1). In total, 272 interviews were completed: 163 with physicians and 109 with nonphysician clinicians and administrators (exhibit 2). Of the physicians and nonphysician clinicians, 58.8 percent were female; clinicians had been in practice for an average of 21.4 years (standard deviation: 11.3; data not shown). Mean interview length was 39.0 minutes (SD: 8.9) (data not shown).

EXHIBIT 1.

Characteristics of participating health systems in the qualitative study of advance care planning codes

	Number
Health systems, total sample	11
Region
West	4
South	2
Midwest	1
Northeast	4
Type
Academic	8
Public	1
Large not-for-profit	2
Religious affiliation noted
Yes	2
No	9
Bed count, mean (SD)	1,947.7 (1,639.2)
Total physician count, mean (SD)	1,884.2 (1,262.0)

SOURCE Authors’ analysis of study data. Health systems data were compiled from the Agency for Healthcare Research and Quality Comparative Health Systems Performance Initiative. Agency for Healthcare Research and Quality. Compendium of U.S. health systems, 2016. Rockville (MD): AHRQ; 2019.

NOTE SD is standard deviation.

EXHIBIT 2.

Sample characteristics of people interviewed in the qualitative study of advance care planning codes, by role and total, 2018–19

Characteristics	Physician	Nonphysician cliniciana	Administrativeb	Total sample
Number	163	75	34	272
Age, mean years	50.1	47.7	55.3	50.0
Sex, %
Male	55.0	16.2	24.0	41.3
Female	45.0	83.8	76.0	58.8
Race, %
White	83.8	91.2	88.0	87.0
Black or African American	2.7	2.9	8.0	2.9
Asian	13.5	5.9	4.0	10.0
Hispanic ethnicity, %	2.0	1.0	4
Languages spoken, %
English only	67.9	84.9	71.4	72.9
Spanish	15.1	4.1	7.1	11.2
Other	17.0	11.0	21.4	15.9
Years in practice, mean	22.3	19.1	20.7	21.4
Region, number
Northeast	59	23	11	93
Northwest	27	11	2	40
South	21	15	8	44
West	56	26	13	95
Specialty, number
Internal medicine/hospitalist	85	2	—c	87
Palliative care/hospice	37	10	—c	47
Primary care	26	7	—c	33
Geriatrics	27	5	—c	32
Surgery	20	0	—c	20
Medical oncology	7	2	—c	9
Neurology	8	1	—c	9
Critical care	7	0	—c	7
Pulmonary	4	1	—c	5
Emergency medicine	3	0	—c	3
Obstetrics-gynecology	2	0	—c	2
Infectious disease	2	0	—c	2
Advanced heart failure or transplant	1	0	—c	1
Cardiology	1	0	—c	1

SOURCE Authors’ analysis of study data.

NOTES N = 272. The number of responses for some characteristics might not add up to the total number of interviewees listed because of some respondents being listed in multiple categories (for example, specialty) or missing responses.

^aIncludes nurse practitioners, physician assistants, registered nurses, and chaplains.

^bIncludes management and billing specialists.

^cNot applicable.

Overall, 37.8 percent of eligible clinicians reported billing regularly for advance care planning, including 13.8 percent of advance practice professionals (data not shown). Most clinicians who billed for ACP used time-based billing codes or chose higher-level evaluation and management codes instead of the designated ACP codes. Clinicians from geriatrics and palliative care departments billed most frequently, at 64.7 percent and 60 percent, respectively (data not shown). Themes included code-based constraints, burdening patients with unexpected charges, ethical concerns with ACP billing, incentives signaling the importance of billing, and workflow burden and the need for institutional support and training (appendix exhibits F and G).¹⁷

Below we summarize responses by theme and subtheme. Where a respondent is quoted, the quotation is followed by a respondent identity code: the respondent number (1–272) and an alphabetic health system code (A–K). Exhibit 3 presents key characteristics of the quoted respondents.

EXHIBIT 3.

Characteristics of quoted respondents in the qualitative study of advance care planning codes, 2018–19

Respondent identifier	Clinical role	Specialty
027B	Physician	Internal medicine
040B	Nonclinical	Billing
051C	Physician	Geriatrics
052C	Physician	Internal medicine
053C	Physician	Palliative care
060C	Physician	Geriatrics
064C	Physician	Primary care
071D	Physician	Internal medicine
072D	Physician	Palliative care
081D	Physician	Geriatrics
082D	Nurse practitioner	Palliative care
085D	Nonclinical	Administration
092D	Physician	Pulmonology
110F	Physician	Palliative care
121F	Physician	Palliative care
132F	Social worker	Internal medicine
141G	Physician	Palliative care
144G	Physician	Internal medicine
146G	Physician	Geriatrics
147G	Nonclinical	Billing
149G	Physician	Surgery
150G	Physician	Palliative care
151G	Physician	Internal medicine
162G	Physician	Internal medicine
164G	Physician	Internal medicine
187H	Physician	Internal medicine
200H	Physician	Palliative care
232K	Physician	Internal medicine
235K	Physician	Primary care
241K	Physician	Palliative care
248K	Nurse practitioner	Administration

SOURCE Authors’ analysis of study data.

NOTE The respondent identifiers refer to the respondent’s number (2–272 out of the 272 responses) and a code for the health system they represent (A–K).

CODE-BASED CONSTRAINTS

▸ ONEROUS TIME REQUIREMENTS:

Many primary care providers felt unable to use ACP codes because of time requirements. Primary care providers described repeated shorter ACP conversations during multiple visits over the course of weeks or months. One explained, “[That] you have to talk for at least fifteen minutes is very constricting because in primary care that just doesn’t happen. I can have a very amazing conversation…a continuation from the last time that lasted five minutes. I can’t bill for it, but it gave me all the information that I needed to document” (053C). Another said, “There are definitely times that we don’t spend sixteen minutes or more…so I don’t use the billing code…even though we are having a really meaningful valuable conversation” (082D).

ACP billing was largely endorsed by palliative care and geriatrics clinicians, who provided ACP services frequently during longer office, inpatient, or facility visits.

▸ INABILITY TO BILL:

Coding constraints exclude valued professionals such as chaplains, nurses, and social workers from being able to use the codes to bill for their services. Participants noted that these clinicians should be able to bill independently for ACP. In some practices this led to division of labor by role and insurance type to optimize efficiency and billing. One social worker explained, “We have a psychologist on our team, so he basically sees any patient that has insurance. So, if you’re uninsured, underinsured, you have some form of Medicaid he doesn’t take, then I’m seeing that patient” (132F). An administrator noted: “We have a lot of people who are not physicians or nurse practitioners who are doing this work and doing it really well, but [cannot] bill” (248K).

▸ EXCLUDING SOME HEALTH SYSTEMS:

The ACP codes largely excluded federally qualified health centers, which are unable to bill for ACP as a separate service because of the nature of their payment system: “The problem is [that the federally qualified health centers are] written into our contracts…. Chronic care management is not included in the [federally qualified health center] services. ACP could be one of them—so could chronic care management—but they are not in our contracts right now” (040B).

BURDEN ON PATIENTS

Clinicians expressed concern that ACP billing would cause unexpected charges, financial burden, and patient dissatisfaction. Outside of wellness visits, clinicians noted that ACP codes would impose additional costs (20 percent of coinsurance) for Medicare patients without supplemental insurance. Unsure of whether patients would incur copays, some clinicians avoided billing entirely. One explained, “My main concern with the codes is having patients get stuck with…an extra copay or an extra charge. I shy away from [billing] because I worry about creating another barrier for [patients] to access support to help them make decisions” (200H). Some clinicians considered patients’ finances and strategically timed ACP to save patients money: “A lot of our patients financially are struggling. …We try to be mindful of the cost and incorporate our ACP in the annual wellness visit” (121F).

ETHICAL CONCERNS

Some clinicians emphasized cultural and ethical concerns regarding ACP billing. One said, “It didn’t feel good to be paid more to do [ACP]. I know it’s part of what a [primary care provider] does, [but] I don’t [get] paid more to have my shared decision-making discussion with you about your mammogram…. I’m just going to put in my preventative health code and not [ACP codes]” (151G). Another explained, “I truly don’t think I have ever coded for [ACP], which…is a good thing. This is being done for the cultural reasons and the patient reasons. …[ACP] has much less do with the monetary rewards and much more for the clinical benefit” (235K). Another offered, “Doctors don’t want to submit a charge for ACP because they feel like [it’s] just part of my package deal” (110F). Clinicians were uncomfortable billing for ACP when it was not the primary appointment goal. One said, “If there were ten other things that happened that visit, I probably don’t [bill]” (162G).

INCENTIVES

Clinicians cited the importance of incentives to their billing decisions. Low reimbursement rate, lack of a relative value unit–based salary structure (especially among palliative care providers), and ACP billing not being an institutional priority were primary barriers. Conversely, institutional prioritization of ACP overcame these barriers and supported billing.

LOW REIMBURSEMENT

Many clinicians identified low payment as a barrier to using ACP CPT codes, instead increasing revenue more efficiently by upcoding visit complexity or billing for time using codes they were accustomed to. One clinician explained, “I can usually bill as some other diagnosis code, and then because of the [ACP] discussions, I can code to a higher level” (081D). Another explained, “I never bill separately from my usual [evaluation and management] bill. It takes time…with all the evidence you need to document…and then bill. I would rather write a high-level summary. …In my world, the trade-off is not really worth it” (144G). Clinicians explained that using established codes was easier and faster: “I get compensated, but whether it’s called ACP or just very long visits [is] not as important to me” (241K).

CLINICIAN PAYMENT STRUCTURE

At some sites, payments to clinicians in some departments, including palliative care, were not relative value unit based, and clinicians were not incentivized to bill. One said, “When it comes to palliative care, I have no [relative value unit] obligations at all, so I don’t ever worry about billing” (072D). This sentiment was supported by many internists. “When you’re in a larger either hospital-based or large group practice, you’re more of a salaried employee, [and billing] doesn’t have that same drive because it doesn’t impact the bottom line” (151G).

INSTITUTIONAL PRIORITIES

Limited administrative oversight and lack of emphasis on ACP codes also diminished clinicians’ enthusiasm for using them. One geriatrician said: “It’s just pure institutional inertia. Nobody has ever emphasized we should be using them. I get no push either negative or positively” (060C). An internist explained: “Even my boss said it really doesn’t matter. Nobody is checking it. No training. No feedback. No nothing. So, why would I? Nobody seems to care” (027B). Within health systems, institutional revenue and payer mix were also implicated. One said, “[X hospital’s] payer mix is better…so they may not be as excited about adding on this code that gives them twenty extra dollars. Revenue seems to drive [billing]” (147G).

Conversely, if respondents’ institutions prioritized and monitored ACP billing, clinicians felt encouraged to use the codes because they perceived their importance (appendix exhibit G).¹⁷ Clinicians using ACP CPT codes were motivated by clear institutional prioritization of ACP billing and financial incentives largely benefiting the institution. One explained: “We chose [an ACP quality metric] because we had a lot of people who thought [ACP] was a good thing to do” (051C). Another explained: “It’s helpful for [palliative care leadership] to advocate for their value.… They really want me to bill ACP as often as I can, not necessarily because they make so much money billing for it, but…it provides verifiable evidence to the Department of Medicine how much work we’re doing” (141G). Others described CMS’s role in determining ACP’s importance: “[ACP is] a big CMS initiative, and we’re a part of [the] Comprehensive Primary Care Initiative, which has had [ACP as] an incentive… align[ed with] reimbursement” (071D).

WORKFLOW BURDEN

Many clinicians cited difficulties using new billing codes and workflow implications. One said, “You do it for a week because it’s fresh in your mind and then you just forget. We’re all set in our ways” (232K). Clinicians described lack of “muscle memory” (150G) associated with billing for ACP, making separate billing not obvious. A hospitalist explained: “[ACP is] sort of part of baked into what we do and it has been…for years. Billing for it is not baked into what we do, it’s a separate thing” (146G).

Many lacked training in using ACP codes: “It’s extremely frustrating, and I have no idea how to bill. …[There are] so many elements that you have to put in to allow you to bill that I don’t bother. There is no easy guide” (164G).

Absence of easily accessible ACP-specific billing functions in the EHR impeded many clinicians, even those eager to use the ACP CPTcodes. One said, “There are too many darn clicks already. …That’s part of the barrier. So you can do time-based billing” (052C). Another echoed: “It’s so completely impractical.We’re strapped as it is. No one has time to be sitting and going to billing codes” (149G). One explained, “I find it annoying…to think about which button to push” (187H).

OVERCOMING BARRIERS

Systems that prioritized ACP billing offered directed training about using ACP CPT codes. Many clinicians described “cheat sheets” with accessible billing guidance, training, and people they could rely on for help (champions). Local champions played an important role, such as having the “ability and interest to go train all of my colleagues and say, ‘Follow me. Here is how you do it right.’ …It takes that investment. We do everything we can to leave no dime behind” (092D).

Health care systems that prioritized ACP billing sometimes included dedicated staff to bill and review notes for additional billing opportunities. “We see sometimes that the documentation is there, but the billing was forgotten. But we have a whole system where our billing team just will flag them and [ensure] that the right code was put in” (085D). Others pointed to EHR investments: “We have incorporated it into the workflow…because our EHR vendor…makes it very easy for billable clinicians to complete all the appropriate elements and then for our professional fee abstractors to identify and bill for those services” (092D). Another approach at another early ACP leader site involved “[creating] our generic smart phrase and our generic billing code phrase” (064C).

Discussion

An aging population requires both growth in capacity for advance care planning and reliable access to it. Although Medicare and most health plans now reimburse for ACP, billing for the service using the codes created for that purpose has not been widely adopted for reasons that have not been well understood.^10,13 Our national qualitative study of 272 clinicians and administrators at eleven diverse health systems found that few clinicians bill for ACP and that few use ACP codes when billing for ACP, consistent with previous studies.^9,12 Using qualitative methods uniquely suited for hypothesis generating, we identified barriers and facilitators to ACP code use at different levels of the health care ecosystem: CMS, institutional, and clinician. At the CMS policy level, barriers included restrictive code constraints and low reimbursement, whereas institutional barriers included insufficient incentives and interruption to clinical workflow. Clinician-level concerns for financially burdening patients and lack of knowledge about ACP codes also impeded billing. The codes offer an opportunity to align clinician reimbursement for providing valuable ACP services consistent with patients’ preferences. Overcoming barriers to their use could encourage health systems and clinicians to ensure adequate and equitable access to ACP.

Low value-to-effort trade-off hindered ACP billing.²¹ Using ACP codes required additional reporting and billing steps, reimbursement was seen as similar to commonly used evaluation and management codes, and many perceived the payment to be too low. Similarly challenging, many clinicians who provided ACP (for example, social workers and nurses) were unable to use the codes. In the primary care setting, time requirements and focus on a single session proved onerous. Primary care providers viewed high-quality ACP as occurring in shorter increments over multiple visits. These barriers can be overcome. For example, ACP codes can be reported along with evaluation and management codes, thus increasing reimbursement substantially.²² Both improving reimbursement for practitioners who inconsistently deliver ACP services and supporting service lines that commonly deliver ACP services are needed to ensure that ACP becomes more universally available. Although barriers to feasibility are great, reimbursing multiple shorter ACP discussions is more consistent with best practices that allow for revisiting preferences and decisions over time with disease progression.²³

Ethical concerns about billing for ACP and financially burdening patients has stymied the use of ACP codes.²⁴ Clinicians viewed ACP as a service already incorporated in their role and were concerned about the appearance of withholding care. Burdening patients with unwanted and unexpected copays further deterred clinicians from billing for ACP.^13,18 Both reimbursing clinicians for ACP and charging copays to patients may be unpopular or poorly understood by patients. For example, Catherine Auriemma and colleagues found that although 90 percent of adults surveyed supported a range of hypothetical interventions to increase ACP, only 23 percent supported physician reimbursement for patients’ completion of advance directives, compared with 58 percent who supported patient reimbursement for completing them.²⁵ Demonstrating ACP’s value to patients and outcomes while eliminating beneficiary cost sharing for ACP services would directly address clinicians’ concerns regarding patient financial burden. Although this would likely require legislation, there is precedent for the elimination of cost sharing for preventive services established in the Affordable Care Act, and congressional legislation (H.R. 4755, the Seniors’ Chronic Care Management Improvement Act of 2021) was reintroduced in January 2021 to eliminate cost sharing for other high-value services such as chronic care management. Importantly, given limited access to ACP for structurally marginalized populations, attention should focus on reducing patient costs and constraints on ACP billing for federally qualified health centers.^26–28

Integrating ACP billing into their workflow proved exceptionally challenging for clinicians, who had little incentive to overcome these barriers. Yet some sites did overcome them. At the institution level we found that many clinicians were motivated by nonmonetary incentives, as prior studies have found, including improving performance on quality measures and enhancing the status of their institution.²⁹ Institutions can support ACP billing by monitoring this measuremore closely, setting benchmarks, engaging local champions, and investing in training and support to ensure that documentation is clear and achievable with minimal disruption to workflow. Also similar to prior findings, we found that investment to minimize EHR demands is crucial.³⁰ CMS could increase incentive payments for ACP in the Quality Payment Program, which would include Alternative Payment Models such as the new Primary Care First and Direct Contracting models, among others. Such policy changes could further incentivize institutions to integrate ACP into their workflow and support clinicians to increase ACP code use.

Conclusion

Limited use of advance care planning CPT codes has, to date, undermined their potential to inform policy. This large, national qualitative study identified facilitators and barriers to use from the provider’s perspective. Improving clinician training about ACP provision and billing, reducing patient cost sharing, and systematic monitoring of ACP and incentives to document and bill may promote greater adoption of ACP codes, improve ease in monitoring of ACP, and increase equitable access to a service that is essential to patient-centered care. ■