Between 2018 and 2023, the number of concierge and direct primary care practices in the United States surged by 83.1 percent, growing from 1,658 to 3,036 practice sites, according to a December 2025 study from Johns Hopkins. The DPC Frontier mapper now lists more than 2,700 active practices across all 50 states. The global direct primary care market was valued at approximately $61 billion in 2024, with projections placing it near $93 billion by 2034. This is no longer a fringe movement. It is an emerging market category with real capital behind it.
But growth without structural discipline creates risk. And the risk that has gone largely unexamined in the DPC advocacy space is the one that matters most to the broader healthcare system: what happens to insurance risk pools when healthy lives migrate out of them?
The structural promise of direct care
The value proposition of direct primary care is real. Patients receive longer consultations, same-day or next-day access, and a relationship with a physician who is not managing a panel of 2,300 patients while navigating prior authorizations. Physicians reclaim clinical autonomy and escape administrative overhead that can consume 10 to 20 percent of every dollar collected in a fee-for-service model. A 2020 Society of Actuaries study found that the introduction of a DPC option in one employer’s self-insured plan was associated with a 12.6 percent reduction in overall demand for healthcare services and a 40.5 percent drop in emergency department usage. Another analysis from the American Consumer Institute reported a 20 percent reduction in hospital admissions and a 41 percent decrease in ER visits under DPC, for memberships as low as $40 per month.
These are meaningful outcomes. They deserve serious attention. But they do not exist in a vacuum. They exist inside a healthcare financing architecture that depends on the distribution of healthy lives across risk pools. And that is where the conversation gets uncomfortable.
The risk pool problem no one wants to name
Consider an employer with 500 employees who introduces a DPC option. The 150 healthiest employees migrate to DPC and downgrade to catastrophic-only coverage. The group plan they left now carries a sicker risk profile, and the insurer adjusts accordingly. That employer just improved primary care access for 150 people and raised premiums for the other 350. This is not a hypothetical edge case. It is the predictable outcome of layering a voluntary membership model on top of a risk-pooled financing system without accounting for selection effects. The World Health Organization has identified fragmented pooling as a direct cause of health system inefficiency, noting that it limits the redistribution of prepaid funds and concentrates financial risk rather than spreading it.
Direct care models, by design, operate outside traditional insurance frameworks. When healthier individuals opt into DPC and drop or downgrade their comprehensive coverage, the insurance pool they leave behind concentrates a higher proportion of individuals with greater healthcare needs. This is adverse selection, and its mechanics are well documented. The American Academy of Actuaries has noted that adverse selection increases premiums for everyone remaining in a health insurance plan because it results in a pool of enrollees with higher-than-average healthcare costs, which can trigger a premium spiral as more healthy individuals exit.
Georgetown University’s Center on Health Insurance Reforms raised this concern directly in 2018, warning that DPC arrangements, particularly after the repeal of the individual mandate penalty, could siphon healthy consumers away from ACA-compliant plans, leaving behind a sicker risk pool and further contributing to premium increases. The Commonwealth Fund echoed this analysis, noting that without regulatory oversight, there is no way to measure the selection effects DPC creates in the broader market. The point is not that DPC practices are deliberately screening for healthy patients. It is that a voluntary, membership-based model will produce selection effects by default, and the impact on the risk pool is the same whether the selection is intentional or not.
The DPC community has largely responded to these concerns with silence or deflection. The standard rebuttal is that DPC is not insurance and therefore does not affect insurance markets. That argument misunderstands how risk pools function. The question is not whether a DPC practice is technically an insurer. The question is whether the migration of healthy lives out of comprehensive coverage into membership-based primary care changes the actuarial profile of the pools they leave behind. The answer, structurally, is yes.
Why this becomes a scaling problem
At 2,700 practices serving an estimated 250,000 patients, DPC’s impact on national risk pools remains marginal. But the trajectory tells a different story. Practice counts have more than doubled since 2019. Corporate-affiliated DPC practices grew by 576 percent during the Johns Hopkins study period. Employer-sponsored memberships now account for 58 percent of all DPC memberships, up 18 points since 2022, according to a 2025 report from Hint Health that compiled data from 2,400 clinicians and 1.2 million members. Thirty-three states have passed laws explicitly defining DPC as not insurance, removing regulatory friction. Policy tailwinds including HSA/HDHP provisions and Oregon’s HB 2540 are lowering remaining barriers.
This is the inflection point. When DPC was a handful of independent practices, adverse selection effects were a rounding error. As the model scales toward mainstream employer adoption and nine-figure market valuations, those effects compound. Each healthy employee who exits a group plan or downgrades to catastrophic-only coverage changes the math for the pool they left. Multiply that across thousands of employers, and the systemic impact becomes material.
Historical precedent reinforces this concern. The Affordable Care Act’s own experience demonstrated that when healthy individuals leave or avoid risk pools, premiums rise, competition narrows, and market instability follows. High-risk pool experiments at the state level have consistently been underfunded and inadequate. The Niskanen Center estimated that covering just the top 5 percent of healthcare spenders would cost approximately $1.5 trillion, roughly 7 percent of GDP. The lesson from every prior attempt at risk segmentation is the same: separating healthy lives from sick ones does not reduce total system cost. It redistributes it, and usually upward.
Designing for integration, not fragmentation
None of this means DPC should be abandoned. It means DPC must be designed to scale without breaking the financing architecture it sits beside. That requires a shift from philosophical advocacy to structural engineering.
Mandate catastrophic pairing. DPC should never function as a substitute for comprehensive coverage. It should function as a layer within a benefits ecosystem that includes catastrophic or high-deductible insurance. The ACA’s Section 1301 already envisions this by allowing DPC to compete on exchanges when paired with a qualifying health plan. Employers and advisors should treat this pairing as non-negotiable, not optional. Every DPC enrollment that displaces rather than supplements insurance coverage is a vote against system stability.
Build risk adjustment into DPC-adjacent plan design. The ACA’s permanent risk adjustment program transfers payments among insurers based on the relative health risk of their enrollees. As DPC scales, regulators and plan designers need to account for the selection effects that DPC creates in the broader market. This could mean adjusting risk corridor calculations to reflect DPC-driven migration patterns or creating reporting requirements that allow actuaries to track the health status of populations entering and exiting DPC arrangements. A 2024 study published in BMC Health Services Research demonstrated that high-risk pooling mechanisms can effectively mitigate selection incentives even in markets with sophisticated but imperfect risk equalization.
Anchor DPC in employer contracting, not consumer retail. The data already points in this direction. Employer-sponsored DPC memberships carry higher retention rates, 85 percent at 12 months, and create natural risk diversification because employer groups reflect a broader cross-section of health status than self-selecting individual consumers. When employers design DPC as infrastructure within their benefits architecture, controlling pricing logic, data governance, and compliance frameworks, the adverse selection risk drops significantly compared to a retail model where individuals self-select based on perceived health status.
Subject DPC to transparency and reporting standards. Thirty-three states have exempted DPC from insurance regulation. That legislative choice has consequences. Without reporting requirements, there is no mechanism for tracking whether DPC practices are, intentionally or not, concentrating healthier populations. Georgetown’s Center on Health Insurance Reforms found that without state oversight, it is impossible to know the extent to which DPC practices might be engaged in health status underwriting. Transparency does not require treating DPC as insurance. It requires treating DPC as a healthcare delivery model with systemic effects that deserve measurement.
The real test of scale
The DPC market is no longer asking whether it can scale. It is asking how. And the answer to that question will determine whether direct care becomes a durable pillar of the healthcare system or an accelerant of the fragmentation that already drives costs upward.
Physicians don’t leave fee-for-service medicine to become complicit in destabilizing insurance markets. They leave because the system is failing patients. But good intentions do not override actuarial math. A model that attracts healthy lives without accounting for what happens to the pools those lives leave behind is not innovative. It is incomplete.
The next phase of healthcare design is not about choosing between direct care and risk pooling. It is about building systems where both can function simultaneously. That requires compliance infrastructure, economic modeling, and a willingness to have the conversations that the DPC advocacy community has avoided.
Innovation that scales at the expense of equity is not reform. It is fragmentation by another name.
Photo: liuzishan, Getty Images
Dana Y. Lujan, MBA, CHFP, CRCR, is founder of Wellthlinks, a healthcare advisory firm that connects providers and employers to design compliant, innovative care models. With more than 20 years of experience in healthcare operations, contracting, and compliance, she has advised health systems, physician groups, and employers on strategies ranging from value-based contracting to direct primary care adoption. Her thought leadership has been published on KevinMD and Medium, where she writes on innovation, compliance, and employer health strategies. She is passionate about building sustainable models that improve access, reduce costs, and strengthen trust between employers, providers, and employees.
This post appears through the MedCity Influencers program. Anyone can publish their perspective on business and innovation in healthcare on MedCity News through MedCity Influencers. Click here to find out how.
