How do you start a direct primary care (DPC / concierge medicine) practice in 2027?

The physician who leaves on a Friday and has no revenue on Monday

Picture a board-certified family medicine physician in a mid-sized city, eleven years into an employed position inside a hospital-owned group. The panel is roughly 2,400 patients. Visit slots are fifteen minutes on paper and seven minutes in practice. Roughly two hours a night go to inbox work, prior authorization forms, and closing notes that exist mainly so a claim can be coded. The compensation formula is RVU-based, which means the only lever the physician controls is volume, and volume is the exact thing making the job unbearable. This is the profile that fills every DPC introductory session: not an entrepreneur looking for a business, but a clinician looking for a way to keep practicing.

The decision looks simple and is not. On the last Friday of employment, the physician has an income of roughly $230,000 a year and a full panel. On the following Monday, in a pure direct primary care model, the income is whatever a handful of early members have agreed to pay — often eight people, sometimes twenty, occasionally zero because the practice has not opened the doors yet. Nothing about the physician's clinical skill changed over that weekend. What changed is that the revenue stopped arriving from a third-party payer on a 30-60-90 day claims cycle and started arriving as monthly subscription charges the physician has to individually earn, one household at a time.

That gap is the actual startup risk, and it is why the capital question is usually asked backwards. New DPC founders fixate on buildout costs — exam tables, a point-of-care analyzer, the lease — when the buildout is genuinely cheap by medical-practice standards. A modest 800-1,500 square foot clinical space with two to four exam rooms runs $15K-$65K in tenant improvements on a $1,200-$3,500 monthly lease, and basic exam equipment lands somewhere in the $15K-$45K band. Total startup capital for a solo de novo practice in a DPC-exemption state typically falls between $80K and $250K. That is a fraction of what a medical spa, a home health agency, or even a traditional insurance-billing primary care practice with payer credentialing and billing infrastructure would require.

The expensive part is time. It commonly takes 3-6 months from entity formation to the first paying member, 12-18 months to reach breakeven, and 18-36 months of organic growth to reach a stabilized panel of 500-800. During that entire ramp the physician needs to eat. Practitioners who make the transition cleanly hold 12-18 months of personal cash reserves or a bridge line before they resign, and they treat that reserve as non-negotiable startup capital — arguably the single largest line item in the whole plan, somewhere between $45K and $185K depending on household burn and geography.

There is a second, quieter version of this scenario that fails less often. An established physician who already owns a practice converts it rather than starting from zero. Conversion costs run lower, maybe $50K-$150K, because the office, the staff, and the patient relationships already exist. Historical conversion experience is that a minority of an existing insurance panel — commonly cited in the 15-30% range — will follow the physician into a membership model. A 2,400-patient panel converting at 20% yields roughly 480 members, which is already inside the target band. The trade is that the physician absorbs a revenue cliff as insurance contracts wind down while membership ramps, and has to have the difficult conversation with 1,900 patients who will not follow.

How the membership mechanism actually works

Strip away the branding and DPC is a subscription business with a medical license attached. The patient pays a flat monthly fee — commonly $50-$150 for an adult, $20-$50 for a child, with family caps around $150-$300 — by automatic ACH debit or recurring card charge. In exchange the member gets unrestricted primary care: 30-60 minute visits, same-day or next-day access, direct phone, text, email, and video access to the physician, after-hours availability, sometimes house calls, an annual comprehensive exam, chronic disease management, and in-office labs and dispensed medications at or near wholesale.

What is absent from that transaction is the entire billing apparatus. No claim submission. No CPT coding. No prior authorization. No MIPS or MACRA reporting. No Medicare Advantage HEDIS or STAR measures. No payer credentialing cycle. No accounts receivable and no 30-60-90 day collection lag. That absence is where the margin comes from, and it is worth being precise about the arithmetic rather than hand-waving at "less overhead."

A traditional insurance-billed primary care practice typically runs overhead in the neighborhood of 55-70% of collections. Billing and coding alone — staff, clearinghouse fees, denial follow-up, contracted billing services — commonly consumes 7-12% of revenue before anyone touches a patient. Strip that layer out, shrink the front-desk function because there are no eligibility checks and no copay collection, and DPC overhead typically lands in the 30-45% range. On $500K of membership revenue, that difference is roughly $125,000 a year that stays with the practice rather than servicing the claims machine.

The panel math is the other half of the mechanism, and it is the part that trips people who model DPC as "the same job with better payment." A traditional physician carries 2,500-3,000 patients because volume drives RVUs. A DPC physician carries 400-800 because the promise is access, and access is a capacity constraint. Take 600 members at $85 a month: that is $612,000 a year of recurring revenue with essentially no collection risk. Now take the same 600 members and consider what they demand. If each member averages 3.5 encounters a year at 40 minutes, that is 2,100 encounters and roughly 1,400 clinical hours before a single administrative task. It fits, but only barely, and only if the panel cap is enforced. Practices that let the panel drift to 1,000 do not fail on revenue; they fail on the promise, and members leave because the same-day access they bought stopped being real.

mermaid flowchart TD A[Physician decides to launch DPC] --> B{State has DPC-exemption law?} B -->|Yes, 36 states| C[Standard membership agreement] B -->|No| D[Counsel-led structuring first] C --> E{Medicare-heavy panel?} D --> E E -->|Yes, opt out| F[File CMS affidavit, 2-year term] E -->|No or low| G[Remain non-participating] F --> H[Private contract at every intake] G --> H H --> I{Employer contracts wanted?} I -->|Yes| J[ERISA structuring: stipend, ICHRA, or TPA] I -->|No| K[Direct-to-consumer only] J --> L[Disclose HSA interaction to employer] K --> M[Disclose HSA interaction to member] </parameter> </invoke>

The HSA question is the structural headwind that no individual practice can solve. IRS Notice 2018-12 treated DPC fees as payment for medical care under Section 213(d), which creates a conflict with the requirement that an HSA-qualified high-deductible plan not cover non-preventive care below the deductible. The practical result is that a member paying a DPC fee may jeopardize HSA contribution eligibility. Legislative fixes under the Primary Care Enhancement Act banner have been introduced repeatedly and passed the House in 2018, but have not been enacted. Workarounds exist and are imperfect: members forgo HSA contributions, or the employer routes reimbursement through an HRA, ICHRA, or FSA rather than an HSA-paired HDHP. This unresolved status is the most-cited reason benefits brokers hesitate to place DPC alongside an HDHP, and it is the difference between DPC as a consumer product and DPC as a mainstream employer benefit.

Employer contracts introduce ERISA, and this is where an otherwise simple practice acquires genuine legal complexity. Self-funded employer plans are federally regulated, and an employer-sponsored DPC arrangement can be pulled into ERISA plan status, bringing summary plan descriptions, Form 5500 filings, claim appeal procedures, and fiduciary duties along with it. The structures counsel typically evaluates are contracting directly with the employee while the employer funds a stipend or HRA, positioning DPC as a supplemental benefit alongside a medical plan rather than as the plan, or integrating the practice as a primary care provider within an existing self-funded plan administered by a TPA. Employer pricing typically runs on a per-employee-per-month basis in the $35-$75 range, which is lower than retail but arrives in blocks of fifty or two hundred lives instead of one at a time.

Adjacent models are worth understanding because prospective members and referral sources conflate them constantly. Classic concierge practices charge an annual retainer, frequently in the $1,500-$2,500 range at the network level and far higher at the luxury end, and continue billing insurance for visits — the retainer buys access, not the care itself. Urgent care is episodic and insurance-billed with no continuity. Telehealth platforms sell convenience rather than relationship. Employer on-site and near-site clinic operators sell to HR departments, not households. Functional and integrative practices layer hormone optimization, extensive panels, and IV therapy at $175-$350 a month. DPC sits at the purest end: flat fee, no billing, small panel, comprehensive longitudinal care. Being able to draw that map in thirty seconds is a sales skill, not a trivia exercise.

Where new practices actually fail

The dominant failure mode is not clinical and not regulatory. It is that panel growth is slower than the founder modeled, and the personal cash reserve runs out at month fourteen with 240 members on the roster — profitable on paper at the practice level, unable to pay a mortgage at the household level. The defense is arithmetic done before resignation: model 20 net new members a month, not 50, and confirm the household survives 18 months at that rate. Twenty a month is achievable with sustained effort. Fifty is what happens in month three when the launch publicity hits and then does not repeat.

The second failure is treating marketing as a launch event. Member acquisition in this model is overwhelmingly relational and cumulative — Chamber of Commerce lunches, employer benefits fairs, talks to small-business owner groups, local podcast and radio appearances, listings in the DPC directories and mappers that prospective members actually search, a website that ranks for the practice's own town plus "direct primary care," and above all referrals from existing members. A practice that does two community events a month for twenty-four months will beat a practice that spent $20,000 on a launch campaign and then went quiet. Budget attention, not just dollars.

Third: panel drift. Growth feels like success, so the cap gets ignored. At 900 members the same-day promise degrades, visits compress toward the fifteen minutes the physician left in order to escape, and churn climbs among exactly the long-tenured members whose lifetime value the model depends on. Enforce the cap. When it fills, open a waitlist and use the waitlist as the business case for a second clinician — that is a far better growth signal than a demand estimate.

Fourth: scope creep. The pull toward hormone optimization, peptides, IV therapy, aesthetic add-ons, and extensive wellness panels is strong because the per-encounter revenue is high and the patients are already there. Each addition blurs the position, invites a different regulatory posture, and risks converting a primary care practice into a wellness clinic that happens to have a membership. Some practices make that choice deliberately and do well. The failure is making it accidentally, one profitable exception at a time.

Fifth: underestimating how hard this is to scale past solo. The industry is roughly 85-90% solo or two-to-three physician practices for structural reasons, not accidental ones. The venture-backed attempt to industrialize membership primary care has a mixed record — Forward Health raised roughly $400M, opened around nineteen locations with automated care pods, and abruptly closed every one in November 2024. One Medical scaled to hundreds of locations and over a million members, was acquired by Amazon for approximately $3.9B in 2023, and does so on a membership-plus-insurance-billing hybrid rather than pure DPC. The pattern is legible: pure DPC scales through physician-owner-operators and small groups; capital-intensive scale tends to require either insurance billing or employer-channel concentration. Plan accordingly, and be skeptical of any plan whose year-three revenue requires you to become a chain.

Sixth: the paperwork that seems boilerplate and is not. Every intake needs a membership agreement that discloses in plain language that this is not insurance and that the member is responsible for catastrophic, specialty, and hospital coverage; a HIPAA notice of privacy practices; a Medicare private contract where applicable; transparent lab and dispensing pricing; consent for secure electronic communication; and records-release for the prior physician. FTC truth-in-advertising standards apply to how membership benefits are described in marketing, which means the website promises must match what the agreement delivers.

A note on the operational spine, since it is worth borrowing thinking from outside medicine here. A DPC practice is running a subscription business, and the disciplines that make subscription businesses work — RevOps disciplines, in the language other industries use — apply directly. Track monthly recurring revenue, net new members, churn by cohort, acquisition cost by channel, and lifetime value, because those five numbers tell you eighteen months earlier than the bank account whether the practice is working. The vendor stack that supports this is mature: a membership platform handling enrollment, billing, and dunning at roughly $99-$299 a month per provider; an EHR built for primary care rather than for coding at $149-$299; and a HIPAA-compliant communication tool for phone, text, and video at $24-$59. That combined $600-$1,500 monthly is the entire back office, which is the whole point of the model.

Related questions

How long until a solo DPC practice breaks even?

Operating breakeven typically lands around 200 members at an $85 average fee. A defensible physician income requires roughly 300-500. Most consistently-marketing solo practices reach that band in month 12-18, with a stabilized 600-800 panel by year two or three.

Do I have to opt out of Medicare?

Not legally, but roughly 70-85% of solo DPC physicians do because it gives the cleanest structure. Opt-out is a two-year filed commitment requiring a private contract with each beneficiary. If Medicare patients are a large share of your target market, model that loss carefully first.

Can employers pay for DPC memberships?

Yes, and it is the main path beyond solo scale, typically at $35-$75 per employee per month. Structuring matters: ERISA can pull a self-funded arrangement into plan status, so most designs route through a stipend, an HRA or ICHRA, or a supplemental-benefit position alongside a medical plan.

What is the difference between DPC and concierge medicine?

Concierge practices charge an annual retainer and still bill insurance for visits — the retainer buys access. DPC charges a flat monthly fee, bills no insurance at all, and caps the panel at 400-800. DPC is generally cheaper for the member and structurally simpler for the practice.

Does a DPC membership replace health insurance?

No, and saying otherwise creates regulatory exposure. DPC covers primary care only. Members still need catastrophic, specialty, hospital, and surgical coverage, which is why the model pairs naturally with a high-deductible plan or a health-sharing arrangement.

FAQ

How much startup capital do I actually need?

Plan on $80K-$250K for a solo practice in a DPC-exemption state, but read that number carefully: the largest component for most founders is not equipment or buildout, it is the 12-18 months of personal cash reserve — commonly $45K-$185K — needed to cover household expenses while the panel builds. Practices that skip that reserve fail at month fourteen with a viable business and an unpaid mortgage.

What should I charge per month?

Most practices land between $75 and $100 for an adult, within an observed $50-$150 range. Band by age since consumption varies, price children in the $20-$50 range, and set a family cap around $150-$300 — the family cap is the highest-leverage pricing decision because it converts entire households instead of individuals. Offer annual prepay at a 10-15% discount for the cash-flow shape during the ramp.

Can my members still use their HSA?

This is unresolved at the federal level. IRS Notice 2018-12 treated DPC fees as payment for medical care, which conflicts with HSA-qualified high-deductible plan rules and may jeopardize contribution eligibility. Legislative fixes have been introduced repeatedly without enactment. Disclose the issue plainly and route employer funding through an HRA, ICHRA, or FSA rather than an HSA pairing.

How many patients can one physician actually handle?

400-800, against 2,500-3,000 in traditional practice. The cap is not arbitrary — it is what makes 30-60 minute visits, same-day access, and direct physician communication physically possible. At 600 members averaging 3.5 encounters a year at 40 minutes, you are already at roughly 1,400 clinical hours before administrative work. Exceed the cap and you break the product you sold.

Do I need a lab and a dispensary?

Neither is mandatory, but both change the member's math. A CLIA waiver enables in-office basics — CBC, metabolic panel, A1c, lipids, urinalysis, strep, flu — and a wholesale reference lab account covers send-outs. Dispensing generics acquired at $1-$8 against $35-$200 retail frequently saves a chronically-medicated member more than the membership costs. Together they contribute maybe 3-8% of revenue and a disproportionate share of referrals.

Should I plan to grow into a multi-location group?

Be skeptical of plans that require it. The industry is roughly 85-90% solo and small-group for structural reasons, and the best-capitalized attempts to industrialize membership primary care have either closed or moved to insurance-billing hybrids. Growing to two to four physicians in one location is a well-trodden path; becoming a chain is a materially different and less proven business.

Sources

• https://www.aafp.org/family-physician/practice-and-career/delivery-payment-models/direct-primary-care.html
• https://www.dpcfrontier.com/
• https://www.cms.gov/medicare/enrollment-renewal/providers-suppliers/opt-out-affidavits
• https://www.irs.gov/pub/irs-drop/n-18-12.pdf
• https://www.dol.gov/general/topic/health-plans/erisa
• https://www.cdc.gov/clia/php/about/index.html
• https://www.kff.org/health-costs/report/employer-health-benefits-annual-survey/
• https://www.ftc.gov/business-guidance/advertising-marketing
• https://www.congress.gov/bill/115th-congress/house-bill/365
• https://www.deadiversion.usdoj.gov/drugreg/index.html

Related on PULSE

• How do you price a subscription service so churn stays under control?
• What does a healthy member acquisition cost to lifetime value ratio look like?
• How do you build a referral engine that compounds without paid advertising?
• What should a small practice track as its core recurring-revenue metrics?
• How do employer benefits brokers actually decide what to put in front of clients?
• When does adding a second provider make sense in a capacity-capped business?