Growth In Number Of Practices And Clinicians Participating In Concierge And Direct Primary Care, 2018–23

Abstract

Primary care clinicians have expressed growing interest in concierge and direct primary care practices, which often feature smaller patient panels and greater clinical autonomy compared with traditional primary care models. We assessed practice and workforce characteristics using a national sample of concierge and direct primary care practices identified through novel linkages of public and proprietary data. From 2018 to 2023, the number of direct primary care and concierge practice sites grew by 83.1 percent and the number of clinicians participating in them by 78.4 percent. The share of clinicians in concierge and direct primary care practices who were physicians declined from 67.3 percent to 59.7 percent, whereas the proportion of advanced practice clinicians increased. Approximately 60 percent of these clinicians participated in Medicare, suggesting concierge or hybrid practice. Independent ownership decreased from 84.0 percent to 59.7 percent, whereas corporate-affiliated practices grew by 576 percent during this period. The growth in these primary care models may offer substantive benefits to patients and clinicians, but it also raises broader questions about changing clinical practice and access to care.

A growing mismatch between demand and supply in the US threatens access to primary care services, which are critical for preventing and detecting disease, managing chronic illness, and navigating the health care system.^1-3 By 2036, the US is projected to face a shortage of up to 86,000 primary care physicians,⁴ driven by high retirement rates, burnout,^5-7 and declining interest among medical trainees in entering the field.³ Clinicians’ concerns about the traditional primary care model include growing administrative and electronic health record burdens,^8,9 rising patient complexity, and low reimbursement compared with other specialties.^10,11

Evolving health care markets have changed primary care practice models in parallel.¹² A trend toward health care corporatization^13,14 and the employment of physicians by large hospitals and health systems¹⁵ has situated primary care within increasingly larger entities. Recent research has focused on these large organizational entities and their effects on primary care delivery and practice.^7,12-14 Emerging primary care models featuring smaller practices and more personalized care, in the form of concierge and direct primary care, have not received the same level of attention.^16,17

Concierge and direct primary care practices share similarities, including smaller patient panel sizes, greater access and availability, and more personalized services for a fee.¹⁷ They also differ in notable ways. Concierge practices charge an annual retainer fee (averaging a few thousand and up to tens of thousands of dollars)¹⁸ and continue to bill insurance on a fee-for-service basis for office visits and procedures. In contrast, direct primary care practices operate entirely outside of insurance, with patients paying a monthly or annual (often smaller) fee in exchange for a set of predetermined services including appointments, basic laboratory services, and sometimes in-office procedures.¹⁹ One widely used definition of direct primary care,¹⁶ adopted by professional groups such as DPC Alliance, requires practices to fulfill three criteria: charge a periodic fee, not bill third parties on a fee-for-service basis, and not implement any per visit charges that are more than the monthly equivalent of the periodic fee. Some physicians, even within the same practice, may adopt characteristics of both the concierge and direct primary care models.

Despite growing awareness of these emerging primary care models,^20,21 evidence on their scope and characteristics is scant. One estimate suggests that direct primary care practices alone served 800,000 patients across 3,500 clinicians as of 2022,¹⁹ although the data sources underpinning these findings are unclear.²² Less is understood about the clinician workforce in these primary care models, which has implications for care access and quality. In theory, clinicians may enjoy greater clinical autonomy, higher income, and improved work-life balance and job satisfaction with lower overhead costs and administrative burdens.²³ Limited studies of patients’ perceptions of these models also cite improved access, shorter wait times, longer visits, and more personalized care.^24-26 Longitudinal evaluations remain lacking.²⁷

Using novel web-based directory data linked to national proprietary clinician-level data, this study addressed this gap by characterizing clinician and practice characteristics in a national sample of concierge and direct primary care practices. Our findings will help establish a foundation for future research to understand practice model changes and their effects on access, quality, and outcomes in an evolving primary care landscape.

Study Data And Methods

Overview

We identified concierge and direct primary care clinicians, along with their locations, using a national online directory. Next, we matched clinicians to National Plan and Provider Enumeration System data to obtain unique National Provider Identifiers (NPIs). We then linked NPIs to 2023 IQVIA OneKey data to identify their practice and clinician characteristics, and we traced these clinicians retrospectively through each year of IQVIA data from 2018 through 2023 to create a longitudinal data set. Our primary sample represents the concierge and direct primary care workforce as it existed in 2024 and does not account for longitudinal changes, including acquisitions or model transitions, unless reflected in current affiliations.

Data

We constructed a novel data set of concierge and direct primary care clinicians by querying FindMyDirectDoctor, a publicly available web-based tool that compiles entries from concierge and direct primary care practices.²⁸ The platform is a free, patient-facing service, although there are optional subscription services for practitioners for paid advertising and patient acquisition. Given that some practices may self-identify as either concierge or direct primary care, we did not distinguish between these models, but we manually verified practices using web-based searches to confirm their inclusion in our analytic sample. Directory data were collected from January through June 2024.

Provider Matching And Sample Construction

We first linked concierge and direct primary care clinicians from the online directory to 2023 National Plan and Provider Enumeration System data, using a fuzzy matching algorithm based on provider name and practice location. Matches were manually verified. In total, 6,017 (87.0 percent) of the 6,914 providers in our data were successfully matched to an active National Plan and Provider Enumeration System profile with an NPI.

Next, we linked these providers to 2023 IQVIA OneKey data, using NPIs (match rate: 73.6 percent). IQVIA OneKey is a proprietary data set that includes independently verified clinician- and practice-level information, such as provider demographics, practice ownership, and corporate affiliations, for approximately 9.7 million US health care professionals. IQVIA OneKey has been widely used in prior research on health care markets, ownership, and practice patterns.^29,30 We classified all clinicians within a given practice as concierge or direct primary care if at least 65 percent of the clinicians at that practice were listed in the original online directory. We selected this threshold by evaluating clinician distributions across practices, to balance sensitivity and specificity in identifying concierge and direct primary care practices while minimizing misclassification.

To refine our analytic sample, we excluded sixty-two clinicians with missing NPIs; we also excluded a total of fifty-one physicians from specific data years if they had graduated from medical school within the previous three years and were thus likely to be residency trainees (a CONSORT diagram showing the sample construction is in the online appendix).³¹ To supplement missing specialty information for advanced practice clinicians, which include nurse practitioners and physician assistants, we used primary specialty categories from the IQVIA data. More than 99 percent of these clinicians had a specialty listed; the thirty-five clinicians with no specialty listed were included in the analytic sample on the basis of their affiliation with a primary care practice.

To build a longitudinal data set for the period 2018–23, we used the IQVIA data to identify the same practices in prior years, using practice name, location, and a unique practice identifier variable. We also used IQVIA’s business ID field to identify overarching “platform” practices that belonged to the same owner but that could have multiple practice-site locations, allowing us to broaden our sample to include clinicians in other sites belonging to the same platform practice. Finally, we used keyword searches (for example, “direct primary care,” “concierge,” and “executive”) within the IQVIA data to flag additional concierge or direct primary care practices that were not in the original online data source. These practices were then manually validated using online web searches of practice websites to confirm that all identified practices were indeed concierge or direct primary care practices and that affiliated clinicians were appropriately categorized. Clinicians who worked at one of the identified practices in any year during the study period were included in the analytic sample. This approach expanded the original list of online directory providers by 42 percent (see the appendix for additional details on sample construction).³¹

Matching To Medicare Files

Because concierge and direct primary care practices can implement hybrid models, distinguishing between these models is challenging. As a proxy, we identified clinicians billing versus not billing Medicare by linking clinician NPIs to publicly available Medicare Physicians and Other Practitioners’ Public Use Files in each year of our study period (see the appendix).³¹ A match to Medicare was a proxy for clinicians in concierge or hybrid practice.

Analysis

The unit of analysis was the clinician-practice-year. We plotted the growth in concierge and direct primary care practices between 2018 and 2023 and specifically examined clinician characteristics, including self-reported age, gender (with values of male, female, and unknown), specialty, and clinician type, as well as changes in practice characteristics, including corporate affiliation, practice size, and geographic distribution. Using the longitudinal structure of our data set, we tracked clinicians’ movement into and out of concierge and direct primary care practices over time, including the practice settings from which new concierge and direct primary care practitioners entered (“entrants”) and the settings to which providers departing such practices exited (“leavers”). We performed Pearson’s chi-square test and Welch’s two-sample -test to assess statistically significant differences between groups. All analyses were conducted using R, version 4.4.2.

This study was approved by the Institutional Review Board at Oregon Health & Science University.

Limitations

This analysis had several notable limitations. First, our data were not a census of all concierge and direct primary care practices; although we used the largest available public directory for identification, multiple sources of data, and manual verification, our estimates were likely a lower bound. Some practices that met the definition of concierge or direct primary care might not have been listed because of a lack of awareness of the directory, voluntary exclusion, or differing interpretations of model definitions. Because we had limited visibility into model transitions at the practice level (for example, an entire practice transitioning from traditional primary care), our data also underestimated internal conversions to concierge or direct primary care. Nonetheless, our approach parallelled those used in investigations of private equity and corporate ownership, and our match rates to the IQVIA OneKey data were in line with prior research using these approaches, with the advantage of recouping additional clinicians not in the original online directory.^29,32-34

Second, we were unable to systematically distinguish between concierge and direct primary care practices on the basis of third-party insurance billing. Although we used Medicare billing as a proxy, practices may self-identify in various ways, and some may operate under hybrid models. For example, although direct primary care is traditionally defined as not billing insurance, some practices may offer insurance-billed services alongside membership fees, aligning more closely with concierge models. In addition, because participation in an online directory is voluntary, there is a potential for selection bias, which may limit generalizability.

Some of the observed growth may reflect improved data ascertainment over time, although it is unlikely that trends were driven solely by changes in data coverage. Finally, we could have missed concierge and direct primary care practices that changed models or closed before 2024. Despite checks to exclude duplicate or obviously inactive listings, some misclassification may have remained, given our reliance on point-in-time practice affiliations.

Study Results

Total concierge and direct primary care practices grew from 1,658 in 2018 to 3,036 in 2023, for an 83.1 percent increase, and the number of concierge and direct primary care clinicians similarly rose from 3,935 in 2018 to 7,021 in 2023, for a 78.4 percent increase. Practice locations were widely distributed across the US, with concentrations in the Northeast and Southeast (exhibit 1). In 2023, states with the highest number of concierge and direct primary care practices were Texas (), California (), Florida (), and Georgia () (appendix exhibit A1).³¹

Exhibit 1.

Practice locations of concierge and direct primary care practices, 2023

SOURCE Authors’ analysis of FindMyDirectDoctor directory data linked to IQVIA OneKey data. NOTE The size of the dots reflects practice size (that is, number of total providers, including MD/DOs [physicians] and advanced practice clinicians).

The majority of practices had fewer than five clinicians, with only 7–9 percent of practices having more than five clinicians during the study period (exhibit 2, appendix exhibit A2).³¹ The shares of practices with one clinician, two to five clinicians, and more than five clinicians remained stable through the study period, as did clinician age and gender distributions. However, physicians (MD/DOs) as a share of all clinicians in concierge and direct primary care practices decreased from 67.3 percent in 2018 to 59.7 percent in 2023, corresponding to an increase in the share of total clinicians accounted for by advanced practice clinicians, from 32.7 percent to 40.3 percent. Likewise, the share of practices that were independently owned declined from 84.0 percent to 59.7 percent, whereas practices affiliated with corporate owners (for-profit firms excluding hospitals and health systems) increased from 152 practices (9.2 percent) in 2018 to 1,027 practices (33.8 percent) in 2023, for growth of 576 percent in the number of practices. Approximately 60 percent of the sample billed Medicare (suggesting clinicians in concierge or hybrid practices), whereas 40 percent did not (suggesting clinicians in direct primary care).

Exhibit 2.

Practice and clinician characteristics of concierge and direct primary care practices in the US, 2018–23

Characteristics	2018	2019	2020	2021	2022	2023
Total no. of practices	1,658	1,844	2,051	2,337	2,638	3,036
Total no. of clinicians	3,935 4,489		5,089	5,820	6,462	7,021
Clinicians per practice
Mean	2.4	2.4	2.5	2.5	2.4	2.3
Median	1	2	2	2	1	1
Maximum	25	30	42	49	67	75
Practice size (no. of clinicians), %
1	51.9	49.9	49.8	49.2	50.3	53.8
2–5	40.5	41.5	41.2	42.3	41.7	39.7
>5	7.6	8.6	8.9	8.5	8.0	6.5
Clinician age, MD/DO only (years)
Mean	52	52	52	52	53	53
Median	52	52	52	52	52	53
Missing	9	13	13	22	37	50
Clinician gender, % female
MD/DO	39.5	41.3	42.4	43.5	44.2	44.6
Advanced practice clinician	85.5	85.7	85.8	85.3	85.3	84.5
Clinician composition, %
MD/DO	67.3	65.9	64.5	62.7	60.9	59.7
Family medicine or adult primary care physician	94.4	95.0	95.2	95.2	95.5	96.2
Obstetrician-gynecologist	1.8	1.5	1.3	1.3	1.1	0.8
Adolescent or pediatric physician	3.8	3.5	3.5	3.6	3.4	3.0
Advanced practice clinician	32.7	34.1	35.5	37.3	39.1	40.3
Practice affiliation, %
Health system or integrated delivery network	6.8	6.2	5.8	5.0	5.5	6.4
Other corporate owned or affiliated	9.2	10.8	20.9	26.9	30.7	33.8
Independent practice	84.0	82.9	73.3	68.1	63.8	59.7
Medicare provider, %
Matched to Medicare	64.2	62.2	59.5	59.0	58.6	59.0
Not matched to Medicare	35.8	37.8	40.5	41.0	41.4	41.0

SOURCE Authors’ analysis of FindMyDirectDoctor directory data linked to IQVIA OneKey and National Plan and Provider Enumeration System data. NOTES Missingness is shown where relevant. Year of birth was not available consistently for advanced practice clinicians (physician assistants and nurse practitioners), and thus age was calculated for MD/DOs (physicians) only. Practice affiliation categories are mutually exclusive—that is, “other corporate owned or affiliated” were neither independent nor affiliated with a health system or integrated delivery network.

Year-on-year growth in concierge and direct primary care clinicians in our sample ranged from 8.7 percent to 14.4 percent (exhibit 3). Among physicians (MD/DOs) newly entering concierge or direct primary care practices after 2018 (exhibit 4), 841 came from a health system or integrated delivery network, 726 came from an independent practice, and 530 came from another corporate-owned or -affiliated practice. Another 655 had no prior employer, of whom 40 percent were within three to nine years of medical school graduation (data not shown). Only 133 physicians entering concierge or direct primary care practices left a government, federally qualified health center, or community health center setting (exhibit 4). In comparison, the advanced practice clinicians entering concierge and direct primary care practices were more likely to be new clinicians with no prior employer (1,139). The majority of clinicians leaving concierge or direct primary care practices transitioned to either health system, corporate, or independent settings.

Exhibit 3.

Clinicians’ participation in, entry into, and exit from concierge and direct primary care practices, 2018–23

SOURCE Authors’ analysis of FindMyDirectDoctor directory data linked to IQVIA OneKey data. NOTES Of the 2,673 clinicians no longer affiliated with a concierge or direct primary care practice during the study period, 1,781 had identifiable subsequent affiliations, 319 remained active in Medicare but could not be linked to practice type, and 573 had no observable activity in either IQVIA or Medicare data. We retained all leavers in the overall exit count. Totals might not equal the prior year plus entries minus exits because of excluded gap years in provider data (additional methodological details are in the appendix; see note 31 in text).

Exhibit 4.

Previous practice locations of clinicians entering and subsequent practice locations of clinicians leaving concierge and direct primary care practices, 2018–23

SOURCE Authors’ analysis of FindMyDirectDoctor directory data linked to IQVIA OneKey data. NOTES “Entry” reflects the earliest observed concierge or direct primary care affiliation, and “exit” reflects the last known affiliation. Clinicians already practicing in concierge and direct primary care models in 2018 were not included as entrants or leavers. Clinicians with only a single year of data (; 9.0%) were excluded from this figure. “Active, unknown” refers to clinicians who were found in Medicare data after leaving a concierge or direct primary care practice, indicating ongoing clinical activity, but whose subsequent practice type could not be determined. FQHC is federally qualified health center. CHC is community health center. MD/DO indicates physicians. APC is advanced practice clinician (nurse practitioner or physician assistant).

Finally, there were differences in some characteristics between entrants and leavers in concierge and direct primary care practices (appendix exhibit A3).³¹ Entrants were slightly more likely to be physicians (52.1 percent) than advanced practice clinicians (47.9 percent), whereas leavers were more evenly split (49.9 percent physicians, 50.1 percent advanced practice clinicians). On average, leavers were older (mean age, 53; SD: 12) than new entrants (mean age, 49; SD: 12).

Discussion

To our knowledge, this was the first descriptive analysis of a large sample of concierge and direct primary care primary care practices. Using national provider-level data, we observed substantial growth in the number of both practices and clinicians participating in concierge and direct primary care models, which has potential ramifications for the workforce and access to care. Between 2018 and 2023, the number of practice sites nationwide grew by 83 percent, whereas the number of clinicians in these practice models increased by 78 percent, with substantial geographic variation. The vast majority of concierge and direct primary care practices remained small, typically employing fewer than five clinicians. However, we observed a substantial shift in practice ownership, with a decline in independent ownership and a rapid rise in corporate-affiliated practices, which grew by 576 percent during the study period.

Concierge and direct primary care practices have emerged in geographically diverse areas across the US,³⁵ and concierge and hybrid models are also increasingly advertised by health systems alongside their traditional primary care offerings.³⁶ A small 2023 survey by the American Academy of Family Physicians, for example, demonstrated that 9 percent of family medicine physicians were direct primary care physicians.²³ Our findings add to this scant literature by quantifying the rapid growth in these practices nationally, suggesting sustained and notable workforce interest in these models of care. Physicians’ interest in concierge and direct primary care practices has been linked to dissatisfaction with traditional primary care settings,²⁴ including well-known administrative burdens associated with practice management, billing, and revenue-cycle management, as well as direct time constraints on patient care.³⁷ Concierge and direct primary care models, in comparison, offer clinicians smaller patient panels, fewer administrative burdens associated with insurance billing, and increased clinical autonomy, which may improve job satisfaction and extend career longevity.^17,24 However, our study also highlighted evolving structural trends, including the increasing presence of corporate ownership,¹³ in a model that was originally conceived of as a means of preserving independent practice. For example, our data included hybrid and corporate-affiliated models, which charge lower membership fees, bill insurance, and operate at scale. These practices may have accounted for part of the observed growth even as they do not fully align with traditional definitions of concierge or direct primary care. Such trends, which could alter some of the elements that most attract clinicians (and patients) to these models, will be important to monitor.

Physicians entering concierge and direct primary care practices came from a variety of traditional primary care settings.

Notably, we found that physicians entering concierge and direct primary care practices came from a variety of traditional primary care settings, with nearly 30 percent of the entering physicians coming from health systems or integrated delivery networks. A smaller but still substantive share came from corporate-owned or -affiliated settings and independent practices. The reasons for these changes are not well understood, although physician surveys suggest that dissatisfaction about reimbursement, clinical autonomy, burnout, and administrative burdens may play a role.^38,39 Conversely, relatively low rates of those in federally qualified health center settings entered concierge and direct primary care practice. Although this may be due to structural barriers to mobility, including loan repayment obligations, mission alignment, or compensation models, it may also reflect the relative durability of the safety-net workforce despite high rates of burnout and staffing challenges. To enhance workforce planning, future research should explore both protective and risk factors for such transitions.

Simultaneously, we found that growth of clinicians in concierge and direct primary care practices was driven as much by advanced practice clinicians as by physicians. In fact, the proportion of advanced practice clinicians grew from 33 percent to 40 percent over the course of the study period, whereas the share of physicians declined (despite growing in absolute counts). This raises additional workforce considerations, which may have implications for care delivery. Advanced practice clinicians were also more likely to be new entrants to the workforce, which could reflect model appeal, differences in hiring patterns, or the availability of positions in these settings. These shifts may help alleviate physician shortages, but they also could contribute to continuing changes in the composition of the primary care workforce.

It is important to note that concierge and direct primary care models are similar in some ways but differ in others.

Finally, it is important to note that concierge and direct primary care models are similar in some ways but differ in others. For example, direct primary care generally operates outside of insurance and is often framed as a model focused on increasing access and affordability. In contrast, concierge medicine operates within the insurance system, in addition to charging membership fees, which may target higher-income patients seeking enhanced services. Distinguishing between these models is challenging. Our linkage to publicly available Medicare files was an imperfect but useful initial approach; we estimated that about 60 percent of the clinicians in our sample were concierge or hybrid based on their presence in Medicare data, with another 40 percent not participating in Medicare and presumably in direct primary care. Research linking these data to more comprehensive administrative claims data may help elucidate important differences between these models.

Our findings provide empirical evidence on these newer primary care models.²⁰ Growing clinician and patient interest in models that emphasize membership-based revenue structures, longer patient visits, and reduced reliance on third-party billing has had direct policy implications. The recently passed One Big Beautiful Bill Act includes provisions that expand the use of health savings accounts to direct primary care, which may increase access. Some states have also enacted specific laws to regulate direct primary care models, although wide variation exists, including the types of providers eligible for these arrangements and the rules regarding services included and fee amounts. As these models expand, empirical evidence is needed to help inform policies aimed at enhancing workforce sustainability and to understand the broader implications of concierge and direct primary care as an alternative to traditional primary care models.

Conclusion

Concierge and direct primary care practices are an increasingly important part of the primary care landscape in the US, which is marked by an expanding gap between the supply of clinicians and the demand for services. These alternative models have attracted an increasing number of clinicians, from physicians leaving traditional practice to newly trained advanced practice clinicians. Our findings around rising corporate ownership and clinician movement have potentially important implications for the nature of care delivered in these emerging models and for the potential to mitigate broader primary care shortages instead of compounding them. Going forward, the impact of concierge and direct primary care models on access to care, quality of care, and the primary care workforce will be vital for patients, clinicians, and policy makers to understand.

TOPICS.

• Primary care

• Medicare

• Access to care

• Physicians

• Systems of care

• Physician practices

• Ownership

• Patient care

• Quality of care

• Private equity