Adoption and Value of the Medicare Annual Wellness Visit: A Mixed-Methods Study

Abstract

Medicare’s Annual Wellness Visit (AWV) was introduced in 2011 to encourage utilization of preventive services, but many clinicians and patients still do not participate in the visit. We qualitatively and quantitatively assess motivations and clinical and financial value of AWVs from a primary care perspective using interviews and Medicare claims from 2012–2019. Primary care providers with the highest acuity patients had AWV utilization rates 11.3 percentage points lower than providers with the lowest acuity patients; utilization rates were 3.8 percentage points lower in rural counties. Adoption was motivated by patient needs and financial incentives. AWVs closed gaps in preventive care, strengthened patient-provider relationships, facilitated advance care planning, and provided an opportunity to improve quality metrics. Overall, the AWV has the potential to increase use of high-value preventive services although not all clinics have an economic incentive to adopt the visit, which may explain some of the variability in utilization rates.

INTRODUCTION

Medicare provides near-universal health care coverage for older adults in the United States. Before the Affordable Care Act (ACA), Medicare beneficiaries were only entitled to one “Welcome to Medicare” visit within the first 12 months of Part B enrollment. This visit includes a brief, focused physical exam, patient’s health and history review, and development of a plan to keep the patient healthy. All subsequent visits were billed as problem-oriented or evaluation and management visits. Notably, fee-for-service (FFS) Medicare, in which most beneficiaries are enrolled, has never covered a comprehensive annual physical exam for older adults.

Evidence favoring preventive medicine (e.g., vaccination, cancer screening) is strong, yet, more than half of adults aged 65 and older were not up to date on recommended preventive care services in 2010 (Centers for Disease Control and Prevention, 2011). Through the ACA, Medicare undertook multiple policy changes expanding prevention benefits, including introducing the Medicare Annual Wellness Visit (AWV). The AWV is not a physical exam, but rather it is a health check-up to focus on overall care plans with emphasis on preventive health care, disease screening, and care coordination (Colburn & Nothelle, 2018). Since 2011, all Part B-covered beneficiaries are entitled to their Welcome to Medicare Visit, plus an AWV at least one year after Medicare enrollment, and annually thereafter, all with no cost-sharing (see Appendix Table A1). Starting in 2016, advance care planning completed during the AWV can receive additional reimbursement (Centers for Medicare and Medicaid Services, 2020). For clinicians, the AWV offers billable time to discuss prevention and health goals with patients and is reimbursed at a higher rate than many established patient visits (e.g., $175.08 initial AWV, HCPCS “G0438” vs. $111.78 established patient visit, Current Procedural Terminology “99214”).

The AWV would appear to fill an important gap in care for older adults, and a growing body of empirical research suggests that AWVs can improve utilization of evidence-based preventive services (Camacho et al., 2017; Ganguli et al., 2019; Jensen et al., 2015; Jiang et al., 2018; Lind et al., 2021; Toseef et al., 2020). However, AWVs were introduced as volume of primary care work outpaces increases in visit lengths (Abbo et al., 2008). Primary care providers (PCPs) may face limited capacity and constrained resources to serve a growing population of older adults, many with chronic and complex health conditions. Uptake of AWVs has been slow and variable, with approximately 25% of eligible beneficiaries utilizing the benefit in 2018 (Ganguli et al., 2017; Morgan et al., 2021). Variability in visit adoption has been noted by geographic region, clinician characteristics, and patient characteristics, suggesting that clinician- and patient-selection factors drive AWV rates. In-depth assessment of factors influencing clinicians to adopt AWVs are scarce, and previous studies are limited to single health care systems (Beran & Craft, 2015; Simpson et al., 2018).

To address these gaps, we describe clinician, practice, and population characteristics associated with AWV adoption and qualitatively explore clinician perceptions about motivation to adopt and value of AWVs. We accomplish these goals through analysis of national Medicare claims summary data and qualitative interviews with PCPs.

NEW CONTRIBUTIONS

The goal of Medicare’s prevention benefit expansion was to improve delivery of evidence-based preventive care (Lesser & Bazemore, 2009). However, AWVs remain under-utilized, and barriers to adoption are not well-documented and may be unobservable in clinical and administrative data sources. AWV utilization patterns are likely driven by factors at the intersection of clinician and patient incentives. In this study, we explain clinician-side variability in AWV adoption by exploring: 1) how and why AWVs are assimilated into clinical practice, 2) perceived clinical and financial value, and 3) characteristics of the patient panel, clinician, and county-level population associated with utilization rates.

We address a gap in the understanding of who is conducting AWVs and why they are delivered (or not) and provide information to improve subsequent policies aimed at increasing preventive service utilization. By asking clinicians about the clinical and financial value of AWVs, and the outcomes they believe would demonstrate AWV effectiveness, we gain a better understanding of the mechanism for small, but significant effects of AWVs on prevention-related outcomes (Beckman et al., 2019; Camacho et al., 2017; Ganguli et al., 2019; Lind et al., 2021; Lissenden & Yao, 2017; Simpson & Kovich, 2019). If the AWV is a benefit in which Medicare intends to continue investing, it is critical to understand what motivates clinicians to adopt the visit and the barriers faced by non-adopters. Addressing these barriers will improve access to AWVs and may help reduce disparities in access and utilization of preventive services.

CONCEPTUAL FRAMEWORK

Informed by literature on dissemination and implementation in health care settings, we classify AWVs as health care innovations: “a novel set of behaviors, routines, and ways of working that are directed at improving health outcomes, administrative efficiency, cost effectiveness, or users’ experience.” (Greenhalgh et al., 2004) We additionally rely on the Consolidated Framework for Implementation Research (CFIR) as an evaluation framework, (Damschroder et al., 2009) acknowledging that implementation in healthcare settings is multi-level. Our assessment of AWV adoption considers internal (e.g., culture, compatibility, available resources) and external contextual factors (e.g., patient needs, external policies and incentives), characteristics of the adopter (both the clinical setting and individual clinicians), and features of the AWV as an innovation itself (e.g., process, complexity). We incorporate factors from several domains of the CFIR as quantitative variables and qualitative codes in this mixed-methods study to understand how these factors influence AWV implementation. (Kirk et al., 2015)

METHODS

We used an explanatory sequential mixed-methods approach (Creswell & Clark, 2017) combining secondary analysis of quantitative Medicare claims data with the primary collection and analysis of qualitative data from clinician interviews. As the analysis progressed, qualitative findings were iteratively incorporated as variables in the quantitative model and additional questions in interviews (Figure 1). At the point of integration, we focused on concordance between qualitative and quantitative findings (Creswell et al., 2011; Fisher & Hamer, 2020) to provide a richer understanding of AWV adoption than could be gleaned from either data source independently, and to address the study objectives using two complementary data sources.

Figure 1.

Joint Display of Explanatory Sequential Mixed-Methods Study Design

Abbreviations: AWV = Medicare Annual Wellness Visit; PCP = primary care provider; SNOCAP = State Network of Colorado Ambulatory Practices & Partners

QUANTITATIVE DATA

We analyzed data from the 2012–2019 Medicare Provider Utilization and Payment Data: Physician and Other Supplier public use files (POSPUF) to quantify factors associated with AWV adoption and to identify potential interviewees in Colorado. The POSPUF includes 100% final-action physician/supplier Part B non-institutional claims for the Medicare FFS population. Data are aggregated by National Provider Identifier (NPI), Healthcare Common Procedure Coding System (HCPCS) code, and place of service. Clinicians who billed more than 10 occurrences (by HCPCS code) in a calendar year are included. Patient panel characteristics were also extracted from the POSPUF. We extracted physician characteristics from Medicare’s Physician Compare database and county-level demographics from the Health Resources and Services Administration Area Health Resources Files (AHRF).

QUANTITATIVE SAMPLE

We used POSPUF files to identify unique clinicians who conducted AWVs. We limited our analysis to primary care physicians from Internal Medicine, Family Medicine, Geriatrics, or General Practice specialties because they performed more than 90% of AWVs.

QUALITATIVE SAMPLE

We identified Colorado-based PCPs from the 2017 POSPUF (most recent year of data when the initial sample was compiled) who billed HCPCS code “G0438” (initial AWV) and/or “G0439” (subsequent AWV). Total Medicare beneficiaries (reported in POSPUF) was used as the denominator to calculate each clinician’s annual AWV rate. Clinicians were matched to practices in the State Network of Colorado Ambulatory Practices & Partners (SNOCAP) practice-based research network (PBRN) using street address from the POSPUF. SNOCAP is a collaborative affiliation of five PBRNs in Colorado that facilitate joint research projects.

PCPs specializing in Internal Medicine, Family Medicine, and/or Geriatrics were eligible to participate in an interview. We recruited physicians (MD and DO) and non-physician providers, including physician assistants, nurse practitioners, and registered nurses since they may also complete AWVs (Medicare Benefit Policy Manual, 2015).

SNOCAP emailed its member practices with qualifying clinicians. Interested clinicians contacted the research team directly. We also used snowball and purposive sampling strategies as data collection progressed. These approaches helped us reach non-adopters and clinicians who did too few visits to appear in the POSPUF (due to total visit volume or patients with Medicare Advantage plans), or who were not practicing in Colorado in 2017. We tried to recruit clinicians with varying AWV rates, urban/rural practice settings, and other characteristics thought to influence utilization (e.g., physician gender, years in practice) (Shenton, 2004).

QUALITATIVE DATA COLLECTION

We designed a semi-structured interview guide to qualitatively explore clinicians’ motivation to adopt the AWV and their perceptions of its clinical and financial value (interview guide available upon request). One author (MKH) conducted interviews from September 2020 to September 2021; all but two were completed by phone (one video call; one email response). Interviews were audio recorded and transcribed verbatim. The interview guide was shortened from 60 to 30 minutes after the eighth interview to mitigate challenges recruiting PCPs for an hour-long interview during the COVID-19 pandemic. Non-adopters completed a 15-minute interview focusing on reasons for not conducting AWVs. Introductory questions established practice and patient panel characteristics, clinic mission and values, and organizational structure to understand the clinical context. The interview explored the: 1) process of conducting AWVs in the interviewee’s clinical setting, 2) rationale for conducting AWVs, specifically focusing on assimilation and implementation processes, and 3) perceived value of AWVs, with an emphasis on patient-outcomes and financial incentives. Data collection ceased after 29 interviews because content saturation was reached (Corbin & Strauss, 2008).

ANALYSIS

Quantitative Analysis

We modeled the total number of AWVs billed annually per physician as a function of physician-, practice-, and county-level characteristics using negative binomial regression to account for overdispersion of the count data. The denominator (i.e., offset) was the number of unique FFS beneficiaries (panel) receiving services from physician i in calendar year t. The final model took the form:

$$\log \left({E[Y}_{it}]\right)={\log \left(panel\right)+\beta }_0+\beta \prime P_i+\delta \prime X_{it}+\gamma \prime C_{it}+{\alpha }_t$$

where P is a vector of fixed physician characteristics (gender and decades in practice as of 2011), X is a vector of time-varying patient panel characteristics expressed as quartiles (average Medicare patient age, proportion of White patients, and average CMS Hierarchical Condition Category [HCC] quartile), and C is a vector of time-varying county-level characteristics (Medicare Advantage enrollment rate quartile and rural indicator). We clustered standard errors at the county level. We report marginal effects expressed as rates, also known as incidence rates in the context of negative binomial models, to facilitate model interpretation.

Qualitative Analysis

Interview data analysis followed a directed content analysis approach (Hsieh & Shannon, 2005). Starting from the three a priori content areas from the interview guide, coding used an emergent approach to emphasize respondent perspectives and de-emphasize researcher speculation. We used Atlas.ti v9 (Berlin, Germany) for data organization and management. After coding was complete, code categories were developed, and themes identified.

Interview participants provided verbal consent to participate. This study was approved as expedited research by the Colorado Multiple Institutional Review Board (#20–0244).

RESULTS

QUANTITATIVE RESULTS

In total, there were 46,614,648 AWVs conducted by 134,761 unique US-based clinicians (primary care physicians, APPs, and specialists) documented in the POSPUF between 2012–2019. The analysis was restricted to 354,557 physician-years of data (Table 1) from the 82,928 unique physicians who conducted 90.5% (n=36,601,256) of all AWVs during the study period. More than 98% of primary care physicians adopting AWVs were from family medicine or internal medicine specialties. About one third of AWV-adopting physicians were female, and more than half had been in clinical practice for at least 20 years. About 9% of AWV-adopting physicians were in rural counties. Medicare Advantage enrollment in the counties where AWV-adopting physicians worked was about 31%, on average.

Table 1.

Sample and Participant Characteristics

	POSPUF 2012–2019, primary care physicians	Interview Sample
Total N	354,557	29
Provider Characteristics
Female	127,843 (36.06%)	14 (48.3%)
Provider Specialty  Family Medicine  Internal Medicine  Geriatrics  General Practice	175,889 (49.61%) 171,773 (48.45%) 1,996 (0.56%) 4,899 (1.38%)	14 (48.3%) 11 (37.9%) 6 (20.7%) --
Non-Physician Provider (e.g., NP, PA)	N/A	4 (13.8%)
Years in practice  <10  10–19  20–29  30+	33,528 (9.46%) 86,006 (24.26%) 111,595 (31.47%) 123,428 (34.81%)	11 (37.9%) 6 (20.7%) 8 (27.6%) 4 (13.8%)
Patient Panel / Practice Characteristics
Average Medicare patient age  <65  65–74  75–84  85+	6,888 (1.94%) 246,057 (69.40%) 101,071 (28.51) 541 (0.15%)	NR
HCC score, mean(SD)	1.22 (0.36)	NR
Race: White, mean % (SD)	81.51% (17.12)	NR
% of panel with Medicare coverage	NR	43.6% (5 – 95%)
Does AWVs	354,557 (100%)	22 (75.9%)
Federally-Qualified Health Center	NR	4 (13.8%)
County Characteristics
Rural, n(%)	30,557 (8.73%)	4 (13.8%)
Medicare Advantage enrollment, mean % (SD)	30.97% (13.0)	NR

NR: Not reported; N/A: Not applicable; SD: standard deviation

Adoption of AWVs increased significantly over time (Figure 2). Between 2012–2019 there were increases in both the number of unique physicians billing AWVs and the average AWV rate per-physician. The number of physicians adopting AWVs grew 109.8%, from 33,791 in 2012 to 70,889 in 2019. The average AWV rate among adopters increased by 55.7%, from 21.9% in 2012 to 34.1% in 2019.

Figure 2.

Changes in AWV adoption, by year 2012–2019

Source: Author’s analysis of Medicare Provider Utilization and Payment Data: POSPUF, 2012–2019. https://www.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trends-and-Reports/Medicare-Provider-Charge-Data/Physician-and-Other-Supplier

Marginal effects from the negative binomial model are reported in Table 2. For each additional decade since medical school graduation, AWV billing was 2 percentage points higher. Patient panel comorbidity burden (measured by average HCC score) was the largest predictor of AWV rate. Physicians whose average patient was in the highest HCC quartile billed 11.2 percentage points fewer AWVs than physicians whose patients were in the lowest quartile. There was a small association between the racial and ethnic composition of a physician’s Medicare panel and their AWV rate. Practicing in a rural county was associated with a nearly 4 percentage point lower AWV rate, compared to urban counties. Higher Medicare Advantage enrollment in the county was also associated with higher AWV rates (all P<0.001). Findings were consistent, both in trends and the impact of covariates, when the model was stratified by geographic region (Appendix Table A2).

Table 2.

Association between physician, panel, and county characteristics and AWV billing rate, 2012–2019

	Marginal effect (SE)
Time trend
2012	[Ref.]
2013	0.017*** (0.001)
2014	0.030*** (0.001)
2015	0.053*** (0.002)
2016	0.074*** (0.002)
2017	0.097*** (0.002)
2018	0.117*** (0.003)
2019	0.141*** (0.003)
Physician Characteristics
Female	0.020*** (0.002)
Practice decades	0.019*** (0.001)
Patient / Panel Characteristics
Average Medicare patient age
Under 65	[Ref.]
65 – 74	0.009 (0.006)
75 – 84	0.060*** (0.007)
85+	−0.039** (0.017)
White/Caucasian, %
Q1: <76.6	[Ref.]
Q2: 76.7 – 87.6	−0.007** (0.002)
Q3: 87.7 – 92.9	−0.002 (0.003)
Q4: >= 92.3	−0.0003 (0.004)
Average HCC score
Q1: < 0.991	[Ref.]
Q2: 0.992 – 1.13	−0.025*** (0.002)
Q3: 1.14 – 1.33	−0.046*** (0.003)
Q4: >= 1.34	−0.112*** (0.004)
County Characteristics
MA enrollment, %
Q1: <21.8	[Ref.]
Q2: 21.9 – 31.6	0.010** (0.004)
Q3: 31.7 – 41.7	0.018** (0.006)
Q4: >= 41.8	0.013** (0.006)
Rural	−0.038*** (0.004)
Sample size (overall)	N=354,557

Note. Results displayed as marginal effects (incidence rates). The standard errors are clustered at the county-level. Robust standard errors are in parentheses. SE = standard error; Q = Quartile; MA = Medicare Advantage

^*p<0.1

^**p<0.05

^***p<0.001

QUALITATIVE RESULTS

We interviewed 29 primary care providers (25 physicians, 4 other clinicians) from diverse clinic settings and geography across Colorado (Table 1). Interviews were audio recorded and lasted 14 – 71 minutes. About half of clinicians were female and over 40% had been in clinical practice for 20 years or more. Clinicians reported that, on average, about half of their patient panel had Medicare coverage of any type (range 5 – 95%). Four clinicians practiced in rural counties, and four other clinicians practiced in Federally Qualified Health Centers (FQHCs). Seven clinicians did not routinely offer AWVs (non-adopters).

Adoption of the AWV was motivated by both patient needs and financial incentives for the clinic. Clinicians articulated that the value of the AWV related to patient-provider relationships, service delivery, and clinic or health care system objectives. Table 3 contains illustrative quotations from the interviews, organized by theme within each of the main topics from the interview guide (motivation, value, and process). The quantitative results aligned to the themes from the interviews also appear in the table.

Table 3.

Interview themes and illustrative quotations

Motivation	Illustrative Quotations from Clinician Interviews
Patient needs	AWVs entice new patients into the clinic: The sweet spot for Medicare annual wellness exams… was to bring people who didn’t have a regular source of care in and given them an opportunity to address screening and wellness care that we know makes a difference. (Physician, adopter) Clinicians offer AWVs to their healthiest patients: One of our senior providers has said, “Why would we do this if we don’t have any well patients?” That’s not entirely true, but we do have a large proportion of our patients that are frail with multiple chronic illnesses. (Physician, non-adopter) Difficult to find time for wellness for high-need patients: Almost uniformly the rule that people with Medicare, just because they tend to be older and more complicated, they’ve got problems that they want dealt with. Those always take precedent and so these other things, you tend to have to nibble at the edges. You have to tack them on and there’s just not space for a dedicated visit, a wellness type visit. (Physician, non-adopter)
Financial incentives	Revenue from AWVs can cross-subsidize other services: I think it’s something that we need to capture better, as a health system, because it’s something that Medicare pays us to do. We can take care of more people better if we don’t leave money on the table with Medicare. (Physician, adopter) From a pay-for-performance standpoint, the health system valued [the AWV] because it meant that we were meeting more of our metrics. (Physician, adopter) AWVs are reimbursement for work already being done: It was the economical [sic] incentive. When you are in a private practice, your profit margin is five percent. I mean, we’re trying to survive. That was an easy fruit that was hanging low to be able to catch. Maybe I could be more… altruistic? That we want improvement of the quality of the care of the patient. You hope that that’s one of the reasons, but it was economical mainly. (Physician, adopter) AWVs are not financially advantageous in all settings: It’s not clear to me there’s a financial benefit. There may be a little bit, but almost all of our visits are level five [highest acuity] return visits. (Physician, non-adopter)
Value	Illustrative Quotations from Clinician Interviews
Strengthening patient-provider relationships	Value of the patient-provider relationship transcends AWVs: The magic of the wellness visit is development of the relationship with the patient, and having the patient understand that you as a practice care about them, and care about their overall health and wellness. The magic of the wellness visit is the time spent, and the caring and passion communicated. (Physician, adopter) It gives a chance to really meet the patient where they’re at, and find out their circumstances, and intervene where we need to… I’d say it’s cherished time to connect with the patients on a different level, where they’re not in pain, or sick, or in fear. (Physician, adopter)
Closing gaps in care	AWVs are unique: I see the annual wellness visit as this is the one time that I have the time to make certain that everything’s complete and accurate. The rest of the time, when you’re coming in, it’s because you’re sick or you’re injured. Right now, you’re well. Let’s get this right. (Nurse, adopter) Standardization eases conversations about sensitive topics: One of the things I like best about it is the structure kind of forces us to talk about advanced directives. If you got patients who have been reluctant to talk about it, or if patients come in with a huge litany of stuff to their regular visits, it’s easy for them to get put onto the back burner, and never get addressed. With the Medicare wellness, it has to be addressed and so it happens. (Physician, adopter) I’m talking with them about the reason to kind of de-escalate… It gives you an opportunity to talk about some transitions in health and health care. Shifting of reasons for doing things. (Physician, adopter)
Meeting metrics	AWVs facilitate meeting clinical quality metrics: What are their actual chronic medical issues? How can we reflect that in the EMR? How can we show that we’re meeting metrics, that we’re taking care of patients, that they’re filling really important medications or that they’ve got their diabetes under control, and even if they don’t have it under control that we’re checking regularly?... The wellness visits are a really important part for capturing all of that. (Physician, adopter) AWVs are a new quality metric: We’re actually having a clinic-wide push. One of our metrics is actually to increase how many Medicare wellness visits we do as a whole, so we are a bit focused on trying to increase the amount that we do in the clinic. (Physician, adopter)
Process	Illustrative Quotations from Clinician Interviews
Navigating Patient Expectations	Patients and clinicians expectations for AWVs differ: I have my agenda, and the patient is coming with their agenda, which is often totally different than mine. [Laughs]… It’s often a negotiation, and, frankly, things like continuing breast cancer screening doesn’t make it on the list for either of us... The things that are covered in the annual wellness visit might not be high on their agenda.” (Physician, non-adopter) Clinicians spend a lot of time explaining the AWV: This has to involve a discussion at every visit between the provider and the patient, because there’s no possible way to screen out and get an “appropriate candidate” for only an Annual Wellness Visit. What we’ve started doing is telling the patients up front, we’re gonna follow-up for your hypertension and other things today, and also do a wellness visit, and so there will be an E&M charge to this, but we’re gonna ask you a bunch of other questions related to screening and everything else… I’ve gotten away from explaining exactly what the visit is and how we can limit it so it’s free. (Physician, adopter)
Workload Re-Distribution	Work redistribution is efficient, but may conflict with relationship building: A lot of the questions, really, a medical assistant could ask, and you could do it in a way that made it more efficient for the practice. (Physician, adopter) If you are having just providers do the annual wellness visit, that’s not a necessarily good use of provider time, versus other members of the healthcare team… But having said that, I think as always, patients prefer to have their providers do as much as they can, one-on-one with the patient. (Physician, adopter)
Practice Culture and Norms	Culture of wellness influences receptivity among clinicians: This is prevention and screening. A lotta doctors say, “Well, that’s not my job. I wanted to take care of sick people.” It’s not always fixin’ people. This is preventing problems and people from breaking. (Physician, adopter) I’ll do them when they’re on my schedule, but I’m not necessarily data mining through my patients to get them in to schedule… I think they are helpful when the patients reach out for them, and I find them to be an endless battle when I am the one reaching out to the patient to have it done. If there’s that external locus and that external motivation, I’m all for it. (Physician, adopter)

Motivation to Adopt the AWV:

Clinicians described AWVs as opportunities to bring patients into the clinic at least once per year. For patients without a usual source of care, clinicians described the benefit of establishing a relationship and getting patients access to evidence-based screening and preventive care, with the added incentive of no cost-sharing for the patient. For established patients, clinicians described the importance of having dedicated time to discuss prevention and long-term planning, topics that are often overlooked during time-constrained visits dominated by acute health issues and chronic disease management.

In addition to the patient incentives of no cost-sharing, clinicians acknowledged their clinic’s financial incentives to adopt the AWV. In some cases, AWVs were an opportunity to generate revenue for uncompensated work that was already done in the clinic. In clinics with an established emphasis on prevention, adopting the AWV ensured they were not leaving reimbursement funds unclaimed. This added revenue stream allowed them to cross-subsidize less profitable, but important, services. In other cases, the AWV created new revenue streams, both from visit reimbursement and when AWVs facilitated meeting metrics associated with pay-for-performance programs. In one safety-net clinic, AWVs were a potential strategy to retain established patients who transitioned from Medicaid to Medicare. Some non-adopters described how AWVs offered little or no financial advantage over the high-acuity visits they routinely billed (e.g., high complexity established patient visit, CPT code 99215). Clinicians not in leadership or decision-making roles in their clinics were more likely to describe patient care-focused rationale for adopting AWVs. Clinicians in decision-making roles about clinic operations and AWVs were more focused on the financial benefits from the visits.

Clinical Value of the AWV:

Clinicians appreciated how AWVs strengthened patient-provider relationships and improved quality of care by providing the opportunity to obtain a holistic view of patients. This benefit transcended the AWV itself, improving care at other encounters through insight into the whole patient and alignment of care with patient wishes and values. The AWV was described as an opportunity to get to know patients outside the context of a problem-oriented visit.

Another benefit of AWVs was the opportunity to close gaps in use of preventive services by setting up referrals that may be overlooked during other time-constrained visits. Many clinics allocate more time for AWVs than regular return patient visits, though this is not standardized across settings. Longer visit lengths for AWVs allowed clinicians to hear and address factors inhibiting completion of some preventive services (e.g., colonoscopy), influence patient behavior, and offered an advantage over trying to address the same visit elements across many encounters in a single year. For these reasons, clinicians agreed with the use of preventive service completion as a measure of AWV effectiveness (e.g., cancer screening, vaccination, abdominal aortic aneurysm screening).

The structure of the AWV, which requires completion of certain elements to qualify for reimbursement from Medicare, induced some clinicians and patients to have conversations about topics that can be difficult to fit into other visits, such as advance care planning. Clinicians explained how AWV structure and standardization may decrease stigma around sensitive topics, as patients may be more likely to engage in sensitive conversations when they are standard and required questions. This allows clinicians to identify situations that can be improved through interventions. Finally, AWVs provide opportunities to discuss needs specific to older patients, like de-escalation of care that may not be beneficial to patients as they age (e.g., stopping cervical cancer screening in women older than 65 years, per USPSTF guidelines).

The AWV provides a structure that allows clinicians to better capture metrics associated with clinical quality measurement. While they may do many elements of the AWV distributed throughout the year, documentation of patient health status, needs, and care delivery improved through the AWV. In other cases, the AWV emerged as a metric itself. Some clinicians described how their health systems, contracted insurance providers, and independent physician/practice associations have internal metrics for AWVs and offer incentives for meeting volume targets, like additional clinical support and bonus payments.

Process and Assimilation into Clinical Practice:

Clinicians described tension between accomplishing the many elements of the AWV and addressing patient needs and expectations. Clinicians reported that patients often expected to have all health care needs addressed during AWVs, including a physical examination. Clinics adopting AWVs used outreach and patient education strategies (e.g., front desk scripts, letters) to set expectations that AWVs are prevention-focused visits that do not address chronic or acute issues and do not include a physical exam. Some clinics used concurrent billing models offering AWV and management visits during the same encounter, with a copay for the management portion only. The absence of a physical exam was cited as a significant deficit in the AWV, and some clinicians believed omitting a physical exam was poor patient care. Many clinicians reported conducting at least a brief physical exam during the AWV. Perceptions that the AWV was complex and required detailed knowledge of which tests and services were covered to qualify as a no-cost visit was reported as a deterrent among non-adopters.

The AWV was profit-generating when clinics modified workflows, built electronic health record templates, and re-distributed some or all AWV work to APPs, nurses, and medical assistants. While some AWV elements can be appropriately delegated to non-physician care team members, doing so could be disruptive to strengthening the patient-provider relationship, which was a key stated value of the AWV. Re-distribution of AWV workload appeared to be more successful in clinics with team-based care models. Profit considerations prompted some clinics to proactively outreach their patients about AWVs to pre-empt receipt of AWVs elsewhere (such as with hospital-based specialist physicians), as explained by one clinician.

Non-Adopters are Deterred by Perceived Complexity and Patient Health Status:

Billing complexity and long visit lengths were reported as barriers by non-adopters. Some smaller practices lacked personnel to delegate AWV tasks to non-physician providers. Some clinicians worried about fitting the numerous AWV elements into the structure of existing visits. One expressed concern that failing to accomplish all the elements of two separately billed visit types (AWV and return patient visit) could be considered fraudulent. Others were amenable to AWVs if the process of co-billing with management visits were clearer and believed offering AWVs may entice healthier patients into the clinic. Still, many non-adopters perceived that their patients were too ill or complex to benefit from a separate “wellness”-focused visit. They were especially sensitive to the burden imposed by an additional visit for frail or complex patients who may face mobility and transportation challenges.

DISCUSSION

To our knowledge, this is the first study to explore adoption and value of the Medicare AWV from the primary care perspective across diverse practice settings. Our analysis of representative Medicare FFS data showed that AWV adoption was slow but increased over time. Most AWVs are conducted by PCPs with healthier patients. Our analysis of interview data confirmed and explained many of the differences we observed in our quantitative models according to physician, patient, and county-level characteristics. Clinicians pointed to prevention-oriented gaps in usual care that were effectively closed through AWVs, including immunizations, referrals to cancer screening, and advance care planning. Clinicians also spoke about the value of having dedicated time to talk with patients, which strengthened patient-provider relationships. Frustration with the rigidity of the AWV was similarly reported by patients in other studies (Beran & Craft, 2015; Solinsky et al., 2021). Despite high reimbursement, the AWV may not be financially advantageous in all primary care settings.

Medicare continues to make large investments into the AWV, with payments exceeding $1.1 billion in 2019 alone (Centers for Medicare and Medicaid Services). This study highlights that AWVs can be a useful tool in primary care settings for delivering evidence-based preventive services, but that the administration of the visit is a significant barrier among adopters and non-adopters alike. Changes to billing requirements that enhance the flexibility of AWVs could improve clinician and patient satisfaction and increase adoption. Dissemination of the health and financial benefits of AWVs to clinicians, patients, and other relevant stakeholders may encourage more engagement with the visit.

One important finding from the interviews is that many factors influencing adoption and utilization of AWVs are not captured using observational characteristics of physicians or patients that are readily available in administrative data sources, such as claims data. Consistent with the broader dissemination and implementation literature (Damschroder et al., 2009; Greenhalgh et al., 2004), we found that the degree of fit with practice norms and culture and clinicians’ own beliefs about the value of AWVs strongly influences adoption. This finding may partially explain variability in adoption described in other studies (Ganguli et al., 2017, 2018; Hu et al., 2015; Lind et al., 2018), and the modest impact of AWVs on prevention-related outcomes when unobserved confounding is not adequately controlled for in the study design (Camacho et al., 2017; Ganguli et al., 2019; Lissenden & Yao, 2017). That physicians serving the highest acuity patients had significantly lower rates of AWV delivery is one example of the patient and provider selection that must be considered in quantitative analyses.

Our findings confirm that, among adopters, the AWV presents an opportunity to increase access and utilization of evidence-based care. Advance care planning (ACP), for instance, may be done as part of a no-cost AWV instead of a separate visit (and can be billed to Medicare in addition to the AWV). Bundling services in this way may help address underutilization of ACP visits (Palmer et al., 2021; Pelland et al., 2019), while simultaneously encouraging reluctant patients to participate in ACP because it is part of the “standard” AWV question set. Similarly, improvements in cancer screening rates may be achieved through the AWV when clinicians recommend that a patient be screened (Klabunde et al., 2006; Lafata et al., 2014) and have dedicated time to address barriers like perceived risk of cancer (Codori et al., 2001) and awareness of services and coverage (Klabunde et al., 2006). These benefits may be especially important in the wake of the COVID-19 pandemic, which disproportionately affected older adults (AWV volume decreased 4.7-percentage points from 2019 to 2020, when AWVs could be done via telehealth or in-person; see Appendix Figure A1). From April to July 2020, a 90.8% decrease in breast cancer screening and a 79.3% decrease in colorectal cancer screening was observed (Chen et al., 2021). The prevention and planning focus of the AWV is suitable for attending to emerging and ongoing sequelae of the pandemic for older adults, such as gaps in preventive care, loneliness, social isolation, and need for advance care planning.

Increasingly, hospitals and for-profit providers are attracted to high AWV reimbursement (Zito & Derricks, 2016), which was reflected in clinicians’ concerns about patient poaching by clinics more aggressively marketing the AWV to attract and retain patients. Continuity and comprehensiveness are core tenets of US-based primary care (O’Malley et al., 2015), and the notion that some patients may exhaust their annual free visit with a clinician other than their PCP is counter to this core mission. The effect of patients receiving AWVs outside of their primary care settings on care fragmentation and outcomes is unknown.

LIMITATIONS

Interpretation of our findings should consider several limitations. In the quantitative analysis we lack representation from clinicians who conduct few AWVs per year because the clinicians must bill at least 11 encounters to appear in the POSPUF. The POSPUF also lacks information about non-physician provider specialty. Because we are unable to distinguish between APPs in primary care and specialty care settings, we omitted them entirely from the quantitative analysis. Next, observations are attributed to the billing provider, even when AWVs are conducted by an RN or medical assistant. APPs and clinical staff are an important part of the AWV workforce, but we were unable to quantify whether they conducted AWVs, as described by some interviewees. Finally, our quantitative analysis does not distinguish characteristics of early vs. late AWV adopters. Insights from the quantitative analysis showed that adoption depends on factors not measured in the POSPUF.

A potential limitation of the qualitative component is that the sample may not be representative. The distribution of characteristics in the interview sample differed from the POSPUF sample. However, because Colorado is demographically and geographically diverse (with urban, rural, and frontier counties) and by recruiting from SNOCAP practices, we benefitted from the variation in size, location, populations served, and organizational structures that make these findings more transferable than research conducted in more uniform practice-based research networks and states. We enhanced the diversity of perspectives by interviewing clinicians conducting AWVs at varying rates, thus improving the generalizability of the findings. Furthermore, Medicare is administered at the federal level and all physicians, regardless of geographic location, are subject to the same Medicare policies. Identification and recruitment of non-adopters was difficult. Perspectives from the limited number of non-adopter participants may not encompass the full scope of reasons clinicians do not perform AWVs. Finally, this study examined Medicare AWV adoption exclusively from the clinician perspective. Patients likely experience different facilitators and barriers to AWV utilization and perceptions of visit value. Exploration and inclusion of patient perspectives will be important for the design of future interventions and policies intended to increase AWV utilization.

CONCLUSION

The AWV has potential to increase use of high-value preventive services when clinicians close gaps in care and clarify patient goals and values. Not all clinics have an economic incentive to adopt the visit, which may partially explain variable utilization rates. Navigating the complex billing and administration requirements is a deterrent for non-adopters. Managing patient expectations about the AWV is a persistent challenge faced by adopting clinicians. Increased flexibility within the AWV could alleviate some administrative burden and improve clinician and patient satisfaction, thereby improving adoption and value of this important benefit.