Direct Primary Care vs. Traditional Health Insurance for Freelancers: The New 2026 HSA Rule

Published: September 23, 2026 · Reading time: 11 min

TL;DR: For 2026, a new rule — IRC §223(c)(1)(E), added by the One Big Beautiful Bill Act (Pub. L. 119-21, §71308) — lets a self-employed person pair a Direct Primary Care (DPC) membership with an HSA-qualified HDHP without the DPC membership blowing up HSA eligibility, as long as the DPC fee stays at or under $150/month self-only or $300/month family (a flat cap through 2026, not indexed until 2027). The catch: a DPC fee isn't insurance, so it can't ride the self-employed health insurance deduction under §162(l) the way the HDHP premium can — but it is a §213(d) "medical care" expense Congress separately added to the HSA's exceptions list, so it can be paid tax-free straight out of the HSA. In this post's worked 2026 example (single filer, $95,000 net Schedule C profit), pairing a $95/month DPC membership with a $340/month HSA-qualified HDHP and a maxed $4,400 HSA beats a comparable $700/month traditional PPO by $2,184.08 on a true after-tax basis — computed from the actual 2026 tax brackets and a full QBI computation, not a flat marginal rate — while also leaving $1,420.00 still growing tax-free inside the HSA. The QBI side has a twist: Option A's qualified-business-income base is $4,320.00 larger, but because the 20%-of-taxable-income cap binds for both options on these facts, that larger base barely changes the allowed deduction.

Every "HDHP vs. PPO" comparison for a freelancer eventually gets to the HSA. This one is different: it's about a third option that didn't have clean tax footing until this year. A Direct Primary Care membership — a flat monthly fee paid straight to a primary-care practice instead of billing insurance per visit — has been growing for a decade, but pairing one with an HSA-qualified HDHP used to sit in a gray zone, because "any non-HDHP plan providing overlapping coverage" is exactly the kind of thing that normally disqualifies HSA eligibility. Congress closed that gap for 2026. Here's what actually changed, what it doesn't change, and a full worked comparison against a traditional plan.

What Changed for 2026: IRC §223(c)(1)(E)

Section 223(c)(1)(A) sets the baseline rule: you're HSA-eligible for a month only if you're covered by an HDHP and not also covered by any other health plan that isn't an HDHP and covers overlapping benefits. A monthly-fee DPC membership is, on its face, exactly that kind of "other" plan.

The One Big Beautiful Bill Act (Pub. L. 119-21, title VII, §71308(a), enacted July 4, 2025) added a new subparagraph (E) to fix this. In the statute's own words, a direct primary care service arrangement:

"shall not be treated as a health plan for purposes of subparagraph (A)(ii)"

— i.e., it's carved out of the disqualifying-coverage test entirely, as long as it fits the definition Congress wrote alongside it. Per §223(c)(1)(E)(ii)(I), that definition requires:

• Medical care (as defined in §213(d)) consisting solely of primary care services
• Provided by primary care practitioners (as defined by reference to the Social Security Act's Medicare provider rules)
• Where the sole compensation is a fixed periodic fee

And per the Limitation in §223(c)(1)(E)(ii)(II), the carve-out only holds if the aggregate DPC fees for that individual, for that month, don't exceed $150 (doubled to $300 for an arrangement covering more than one person). Go over that line — even by combining two separate DPC memberships for the same person — and the whole arrangement falls back out of the definition, reverting to ordinary disqualifying coverage.

This applies to months beginning after December 31, 2025 (§71308(d)) — so for a calendar-year plan, it's a genuinely new-for-2026 rule, not something that's been quietly available for years.

The $150/$300 cap is flat for 2026 — it isn't indexed yet

Most HSA-related dollar figures get a cost-of-living adjustment every year. This one doesn't, not yet. Section 223(g)(1) says each dollar amount in the listed subsections "shall be increased" by the cost-of-living adjustment, but it only reaches the DPC threshold "in the case of taxable years beginning after 2026, (c)(1)(E)(ii)(II)" — and once it does apply, the calculation substitutes "calendar year 2025" as the base year. In plain terms: $150/$300 is the exact number for 2026, full stop, and the first inflation bump doesn't arrive until 2027.

What a DPC membership excludes — and what it doesn't disqualify

Per §223(c)(1)(E)(iii), "primary care services" for this rule specifically excludes:

• Procedures requiring general anesthesia
• Prescription drugs, other than vaccines
• Laboratory services not typically administered in an ambulatory primary-care setting

None of those exclusions disqualify the arrangement — they just mean a DPC membership was never meant to replace real insurance for surgery, specialist drugs, or hospital-grade labs. That's exactly why the statute's design pairs a DPC membership with a genuine HDHP, not with nothing.

The Tax Asymmetry: "Insurance" vs. "Medical Care"

The DPC carve-out solves the HSA-eligibility problem. It does not make a DPC fee insurance, and that distinction drives two separate, opposite results:

So the DPC fee's only real path to a tax benefit is the HSA. If you're pairing a DPC membership with an HDHP specifically to keep the arrangement HSA-eligible, but then paying the DPC fee out of your checking account instead of the HSA, you're leaving that tax benefit on the table for no reason.

Worked Comparison: A Freelancer's Full 2026 Year

Facts: a single-filer freelance consultant, $95,000 net Schedule C profit for 2026, no other income, taking the standard deduction, self-only coverage. One moderate-complexity referral (say, an orthopedic consult and imaging) generates $2,400 in specialist and diagnostic bills during the year — otherwise, routine primary care is used as needed. Premium and fee figures are illustrative market examples for the purpose of the comparison, not quoted rates — get an actual quote for your state, age, and household.

Option A — DPC + HSA-qualified HDHP + HSA:

• DPC membership: $95/month ($1,140/year) — under the $150/month 2026 cap
• HDHP premium: $340/month ($4,080/year); deductible $1,700 (2026 self-only minimum, Rev. Proc. 2025-19), 20% coinsurance after
• HSA contribution: $4,400 (2026 self-only maximum)

Option B — Traditional low-deductible PPO:

• PPO premium: $700/month ($8,400/year); primary care copay $35/visit (4 visits/year); deductible $500, 10% coinsurance after
• Not HSA-eligible — $500 deductible fails the $1,700 2026 HDHP minimum

node -e "
const fmt = n => n.toLocaleString('en-US', {minimumFractionDigits:2, maximumFractionDigits:2});

// 2026 single-filer bracket table, Rev. Proc. 2025-32 (IRB 2025-45 Table 3)
function tax2026Single(ti) {
  if (ti <= 0) return 0;
  const brackets = [
    [0, 12400, 0.10, 0],
    [12400, 50400, 0.12, 1240],
    [50400, 105700, 0.22, 5800],
    [105700, 201775, 0.24, 17966],
    [201775, 256225, 0.32, 41024],
    [256225, 640600, 0.35, 58448],
  ];
  for (const [lo, hi, rate, base] of brackets) {
    if (ti > lo && ti <= hi) return base + rate * (ti - lo);
  }
  return 192979.25 + 0.37 * (ti - 640600);
}

const netProfit = 95000;
const standardDeduction = 16100; // 2026 single, Rev. Proc. 2025-32 sec 4.14

const seTaxable = netProfit * 0.9235;
const seTax = seTaxable * 0.153;
const halfSeTax = seTax / 2; // identical under both options -- health-plan choice never touches Schedule C

// --- Option A: DPC + HSA-qualified HDHP + HSA ---
const dpcAnnual = 95 * 12;
const hdhpPremiumAnnual = 340 * 12;
const hdhpDeductible = 1700; // 2026 self-only HDHP minimum, Rev. Proc. 2025-19
const specialistBill = 2400;
const hdhpOOP = hdhpDeductible + (specialistBill - hdhpDeductible) * 0.20;
const hsaContribution = 4400; // 2026 self-only HSA max, Rev. Proc. 2025-19
const hsaSpentThisYear = dpcAnnual + hdhpOOP;
const hsaLeftInvested = hsaContribution - hsaSpentThisYear;
const genuineSpendA = dpcAnnual + hdhpPremiumAnnual + hdhpOOP;
const sec162lDeductionA = hdhpPremiumAnnual; // DPC fee is NOT insurance -- not Sec 162(l) eligible
const hsaDeductionA = hsaContribution;

// --- Option B: traditional low-deductible PPO ---
const ppoPremiumAnnual = 700 * 12;
const ppoCopays = 35 * 4;
const ppoDeductible = 500;
const ppoSpecialistOOP = ppoDeductible + (specialistBill - ppoDeductible) * 0.10;
const ppoOOP = ppoCopays + ppoSpecialistOOP;
const genuineSpendB = ppoPremiumAnnual + ppoOOP;
const sec162lDeductionB = ppoPremiumAnnual; // the whole premium is insurance
const hsaDeductionB = 0; // \$500 deductible fails the \$1,700 2026 HDHP minimum

// Treas. Reg. Sec 1.199A-3(b)(1)(vi): Sec 162(l) reduces QBI; the HSA deduction (Sec 223) does not
function qbiFigures(sec162l, hsaDed) {
  const qbi = netProfit - halfSeTax - sec162l; // QBI base -- HSA deduction is not attributable to the trade/business, so it never enters here
  const tentative20 = 0.20 * qbi;
  // Sec 199A(e)(1): the taxable-income cap uses taxable income computed WITHOUT the QBI deduction
  // itself, but AFTER every other above-the-line deduction (including the HSA deduction) and the
  // standard deduction -- this is a different number from the QBI base above
  const tiBeforeQBI = netProfit - halfSeTax - sec162l - hsaDed - standardDeduction;
  const taxableIncomeCap = Math.max(0, 0.20 * tiBeforeQBI);
  const qbiDeduction = Math.min(tentative20, taxableIncomeCap); // taxable income cap binds for both options below
  const finalTI = tiBeforeQBI - qbiDeduction;
  return { qbi, tentative20, tiBeforeQBI, taxableIncomeCap, qbiDeduction, finalTI };
}
const qA = qbiFigures(sec162lDeductionA, hsaDeductionA);
const qB = qbiFigures(sec162lDeductionB, hsaDeductionB);
// Below the \$201,750 single 2026 Sec 199A threshold, neither the SSTB phase-out nor the
// W-2-wage/2.5%-property cap can engage regardless of profession -- only the taxable-income cap can bind
const taxA = tax2026Single(qA.finalTI);
const taxB = tax2026Single(qB.finalTI);
const baseline = qbiFigures(0, 0); // no health-related deductions at all, for an apples-to-apples tax-benefit baseline
const taxNoHealth = tax2026Single(baseline.finalTI);
const healthTaxBenefitA = taxNoHealth - taxA;
const healthTaxBenefitB = taxNoHealth - taxB;

const trueNetCostA = genuineSpendA - healthTaxBenefitA;
const trueNetCostB = genuineSpendB - healthTaxBenefitB;

console.log('halfSeTax', fmt(halfSeTax));
console.log('genuineSpendA (DPC + HDHP premium + HDHP OOP)', fmt(genuineSpendA));
console.log('hsaLeftInvested (still an asset, not a cost)', fmt(hsaLeftInvested));
console.log('QBI base (A)', fmt(qA.qbi), '| tentative 20%:', fmt(qA.tentative20));
console.log('taxableIncomeBeforeQBI (A)', fmt(qA.tiBeforeQBI), '| 20%-of-TI cap:', fmt(qA.taxableIncomeCap));
console.log('qbiDeductionAllowed (A)', fmt(qA.qbiDeduction));
console.log('finalTaxableIncomeA', fmt(qA.finalTI));
console.log('healthTaxBenefitA', fmt(healthTaxBenefitA));
console.log('trueNetCostA', fmt(trueNetCostA));
console.log('genuineSpendB (PPO premium + OOP)', fmt(genuineSpendB));
console.log('QBI base (B)', fmt(qB.qbi), '| tentative 20%:', fmt(qB.tentative20));
console.log('taxableIncomeBeforeQBI (B)', fmt(qB.tiBeforeQBI), '| 20%-of-TI cap:', fmt(qB.taxableIncomeCap));
console.log('qbiDeductionAllowed (B)', fmt(qB.qbiDeduction));
console.log('finalTaxableIncomeB', fmt(qB.finalTI));
console.log('healthTaxBenefitB', fmt(healthTaxBenefitB));
console.log('trueNetCostB', fmt(trueNetCostB));
console.log('true net cost gap (B - A)', fmt(trueNetCostB - trueNetCostA));
console.log('QBI base gap (A base minus B base)', fmt(qA.qbi - qB.qbi));
console.log('actual QBI deduction gap (A minus B)', fmt(qA.qbiDeduction - qB.qbiDeduction));
"

Output:

halfSeTax 6,711.54
genuineSpendA (DPC + HDHP premium + HDHP OOP) 7,060.00
hsaLeftInvested (still an asset, not a cost) 1,420.00
QBI base (A) 84,208.46 | tentative 20%: 16,841.69
taxableIncomeBeforeQBI (A) 63,708.46 | 20%-of-TI cap: 12,741.69
qbiDeductionAllowed (A) 12,741.69
finalTaxableIncomeA 50,966.77
healthTaxBenefitA 1,492.48
trueNetCostA 5,567.52
genuineSpendB (PPO premium + OOP) 9,230.00
QBI base (B) 79,888.46 | tentative 20%: 15,977.69
taxableIncomeBeforeQBI (B) 63,788.46 | 20%-of-TI cap: 12,757.69
qbiDeductionAllowed (B) 12,757.69
finalTaxableIncomeB 51,030.77
healthTaxBenefitB 1,478.40
trueNetCostB 7,751.60
true net cost gap (B - A) 2,184.08
QBI base gap (A base minus B base) 4,320.00
actual QBI deduction gap (A minus B) -16.00

Both final-taxable-income figures land squarely inside the 2026 single 22% bracket ($50,400–$105,700, Rev. Proc. 2025-32 Table 3) — this comparison never straddles a bracket boundary, so the tax figures above come from the actual bracket formula, not an assumed flat rate. Neither option's taxable income before QBI comes anywhere close to the $201,750 single §199A threshold, so the SSTB phase-out and the W-2-wage/2.5%-property cap are both irrelevant here regardless of whether the freelancer's specific profession is an SSTB — the only limit that can bind at this income level is the 20%-of-taxable-income cap (§199A(a)(2)), and it binds for both options: each one's tentative 20%-of-QBI figure ($16,841.69 and $15,977.69) is well above its own 20%-of-taxable-income cap ($12,741.69 and $12,757.69), so the cap — not the QBI base — is what actually limits the deduction in both cases.

The DPC-plus-HDHP-plus-HSA combination wins by $2,184.08 here — and that's before counting the $1,420.00 still sitting in the HSA, untouched, continuing to compound tax-free for a future year. Notice that the tax-benefit line is close ($1,492.48 vs. $1,478.40): Option B's bigger premium generates a bigger §162(l) deduction and a bigger raw QBI base, but because the taxable-income cap binds for both options, that bigger base barely matters — Option A's allowed QBI deduction actually comes out $16.00 smaller, not larger, because Option A's larger total above-the-line deductions (§162(l) plus the HSA contribution) leave it with slightly less taxable income to apply the 20% cap against. The real win here is the lower raw cash spend, not a bigger tax deduction — the same lesson the site's HDHP-vs-PPO comparison draws for the plain HDHP-vs-PPO choice.

The QBI Wrinkle

Neither option touches Schedule C or self-employment tax — both the §162(l) deduction and the HSA deduction are personal, above-the-line items on Schedule 1. But they aren't treated identically for the Qualified Business Income deduction. Treas. Reg. §1.199A-3(b)(1)(vi) names several above-the-line deductions "considered attributable to a trade or business" for this purpose — introduced with "deductions such as," so the list is illustrative rather than closed — and includes the deductible portion of self-employment tax under §164(f) and the self-employed health insurance deduction under §162(l) by name. The HSA deduction under §223 is not named, and unlike §162(l) it isn't computed from the business's own income or tied to carrying on the trade or business at all — it's available to any HSA-eligible taxpayer regardless of self-employment — which is the same reasoning the regulation uses for the items it does include, so the omission reads as a considered one rather than an oversight.

That means Option B's $8,400.00 §162(l) deduction reduces the QBI base by $8,400.00, while Option A's $4,080.00 §162(l) deduction reduces it by only $4,080.00 — Option A's QBI base ($84,208.46) really is $4,320.00 larger than Option B's ($79,888.46). §199A has three independent limits, though, and checking only the base's size stops one step too early. Below the $201,750 single 2026 threshold (Rev. Proc. 2025-32 §4.26), the SSTB phase-out and the W-2-wage/2.5%-property cap can't engage at all, whatever the freelancer's profession — the only limit that can bind here is the 20%-of-taxable-income cap under §199A(a)(2). And on this example's own facts, it does bind, for both options: each option's 20%-of-QBI-base figure is well above its own 20%-of-taxable-income ceiling, so the ceiling — not the base — sets the actual deduction. Because Option A's bigger total above-the-line deductions (the §162(l) premium plus the HSA contribution) also leave it with slightly less taxable income than Option B to apply that 20% cap against, the $4,320.00 base advantage doesn't survive the cap: Option A's allowed deduction ($12,741.69) ends up $16.00 smaller than Option B's ($12,757.69), not larger — a rounding error next to the $2,184.08 the raw cash-spend difference is actually worth. The lesson generalizes: a bigger QBI base only helps if the taxable-income cap isn't the binding constraint — check which of the three limits actually applies to your own numbers (the QBI deduction guide walks through all three) before assuming a larger deduction follows from a larger base.

Where a Traditional Plan Still Wins

1. No DPC practice near you, or the local fee is well above $150/month. DPC availability is geographically uneven, and a rural or high-cost-of-living area may not have an option under the statutory cap at all.
2. A genuinely heavy medical year. This example used a moderate $2,400 specialist bill. A year that pushes toward the HDHP's $8,500 out-of-pocket maximum (2026 self-only, Rev. Proc. 2025-19) costs meaningfully more than a PPO with a lower out-of-pocket ceiling — run your own expected worst case, not just a typical year.
3. You need a specialist-heavy network the DPC practice can't quarterback as well as your current PPO's referral system, or you're mid-treatment with specialists your PPO already covers in-network.
4. You won't actually fund the HSA. Without HSA dollars behind it, the DPC fee gets zero tax benefit (see the tax-asymmetry table above) — the entire tax case for this structure depends on pairing it with real HSA contributions, not just the eligibility carve-out.

Audit Triggers & Common Mistakes

1. Claiming the DPC membership fee on the self-employed health insurance deduction (Schedule 1, Line 17). It isn't insurance under §162(l) — only the HDHP or PPO premium belongs there.
2. Stacking two DPC memberships for the same person without checking the combined fee against the $150/month cap. The limitation in §223(c)(1)(E)(ii)(II) aggregates all DPC arrangements covering that individual.
3. Assuming the DPC-HSA carve-out applies to a plan year that started before 2026. §71308(d) applies the rule to months beginning after December 31, 2025 — not retroactively.
4. Pairing a DPC membership with a plan that doesn't actually meet the 2026 HDHP minimums ($1,700/$3,400 deductible, $8,500/$17,000 out-of-pocket) and assuming HSA eligibility anyway. The DPC carve-out removes one obstacle to HSA eligibility; it doesn't waive the HDHP test itself.
5. Paying the DPC fee out of a personal checking account instead of the HSA. That's the entire tax benefit of pairing the two — skipping it converts a tax-free medical expense into an after-tax one for no reason.
6. Assuming the $150/$300 monthly cap is inflation-adjusted for 2026. It isn't — per §223(g)(1), the first cost-of-living bump applies only to taxable years beginning after 2026.

How CentSense Helps

CentSense logs every DPC membership payment, HDHP premium, and out-of-pocket medical receipt as it happens, so you're never reconstructing which dollars came from the HSA and which came from checking at tax time:

• Scan the DPC membership invoice and HDHP premium statement with AI and tag each to the right Schedule 1 line — health insurance deduction versus a non-deductible personal payment
• Track HSA-funded payments separately from out-of-pocket ones, so you can prove which medical expenses were reimbursed tax-free if the IRS ever asks
• Keep a running total of HSA contributions against the 2026 $4,400/$8,750 limits so you don't over-contribute
• Export a CPA-ready category breakdown as CSV when it's time to file

For related reading, see HSA for Freelancers, HSA vs. FSA for the Self-Employed, and HDHP vs. PPO for Freelancers.

Authoritative References

• 26 U.S.C. §223 — Health savings accounts (Cornell LII) — including the new §223(c)(1)(E) direct-primary-care carve-out and the §223(g) cost-of-living escalator
• 26 U.S.C. §162 — Trade or business expenses, subsection (l): self-employed health insurance deduction (Cornell LII)
• 26 U.S.C. §213 — Medical, dental, etc., expenses (Cornell LII)
• 26 CFR §1.199A-3 — Qualified business income, qualified REIT dividends, and qualified PTP income (Cornell LII)
• IRS — Revenue Procedure 2025-19: 2026 inflation-adjusted HSA and HDHP figures
• IRS — Internal Revenue Bulletin 2025-45 (Rev. Proc. 2025-32, §4.01 rate tables, §4.14 standard deduction)
• IRS — Publication 969, Health Savings Accounts and Other Tax-Favored Health Plans
• Pub. L. 119-21 (One Big Beautiful Bill Act), title VII, §71308(a)–(d), July 4, 2025, 139 Stat. 325–326 — the enacting provision for IRC §223(c)(1)(E), codified in the historical notes to 26 U.S.C. §223

Comparing a Direct Primary Care membership against a traditional plan means tracking dollars from three different sources — checking, HSA, and payroll-style premium — without losing track of which is which. Start a free CentSense account, scan every DPC invoice, premium statement, and HSA-funded receipt with AI as it happens, and export a CPA-ready breakdown before you file. Free tier includes 10 AI scans per month.

This guide is general education for U.S. self-employed freelancers evaluating health-coverage structures for the 2026 tax year. It is not personalized tax or insurance advice, and it is not a substitute for confirming your own HDHP's compliance with the 2026 minimums or your own DPC arrangement's compliance with the $150/$300 monthly cap. Premium, membership-fee, and usage figures in the worked example are illustrative, not quoted rates — get actual quotes for your age, state, and household, and confirm HSA eligibility directly with your HSA custodian before contributing. Consult a CPA or EA for your situation.