What else to borrow

This is a beta: early, unfinished, and open to being wrong. If you think something here is wrong or would fail in practice, tell us.

This is an appendix to the full plan. It is the detail behind one part of it, kept whole for readers who want the argument rather than the conclusion.

(cut) Every item named below is already working somewhere. Everything in this Part is recalled from training data and unverified; treat the whole section as a research list rather than as confirmed facts. Items already built into earlier Parts are marked (in).

Setting the total, and making the budget hard

• Taiwan: the floating point value. Almost unknown here. Taiwan's National Health Insurance sets a total global budget for a given period; providers bill in points rather than dollars; at the end of the period the budget is divided by the total points claimed across all providers, so if the system as a whole bills more volume, each point converts to fewer dollars. The budget becomes physically impossible to overrun. Providers collectively police each other's volume, because one hospital's excess dilutes everyone's payment. Taiwan runs a comprehensive single-payer system at a fraction of American cost. As a complementary mechanism: global budgets as described in the payer section of the plan cap each hospital's revenue individually, but if aggregate billing across all hospitals exceeds the national target, the individual caps have leaked. Taiwan's floating point value caps total system spending automatically, catching any such overshoot because the per-point payment rate falls until spending fits the envelope.
• France: ONDAM. Parliament votes an explicit national health spending target every year. It makes the total a democratic decision made in public. "We decided to spend this much on health" becomes a claim democratic politics can argue over, instead of the total emerging as a residue of millions of private transactions that no one voted for. Adopt this.
• Japan: the schedule can go down. Chuikyo has repeatedly cut fees in absolute terms to hold national spending flat. American policy has no vocabulary for a price reduction and needs one. (partly in, the payer section of the plan)
• Germany, the Joint Federal Committee (G-BA). Payers, providers, and patient representatives jointly decide what's covered, with legal force. A better model for a benefits body than a purely technocratic agency: the people who have to live with the decision sit in the room.

Drugs — three mechanisms better than pure negotiation

• Germany, AMNOG. The most elegant drug pricing rule anywhere. A new drug launches at whatever price the manufacturer wants for the first year, during which an independent body assesses whether it is actually better than existing therapy. Price is then negotiated on the basis of that assessment, and if it shows no added benefit, it gets reference-priced to the old drug. Nobody is denied a new medicine; the public does not pay a premium for novelty that isn't an improvement. Adopt this outright. It answers the innovation objection to price-setting.
• New Zealand, PHARMAC. A hard national drug budget and a single tenderer that runs manufacturers against each other, funding one option per therapeutic class. The lowest drug prices in the developed world. The trade-off is slower access to some new drugs, which is a live political controversy in New Zealand. Take the tendering; consider not taking the hard cap.
• Reference pricing. Pay one price for therapeutically equivalent drugs and let the manufacturers compete down to it. Mechanically simple and well proven, it complements the prize fund in the drugs section of the plan: the prize fund covers new medicines whose patents are surrendered, while reference pricing controls the cost of older drug classes still under patent or with multiple therapeutic equivalents on the market.

Primary care — the part America has never built

The plan was thinnest here.

• Costa Rica, EBAIS teams. The most transferable model on this list. Every citizen is assigned to a geographic primary care team (physician, nurse, community health worker) responsible for a defined population of a few thousand people, doing home visits and tracking who is sick before they present. Costa Rica achieves near-American life expectancy on a small fraction of the spending. Community health centers should be organized this way, in rural counties and underserved urban neighborhoods alike.
• Brazil, the Family Health Strategy and its community health agents. Hundreds of thousands of paid community health workers making monthly household visits, with large documented reductions in infant and cardiovascular mortality. Community health work, paid, structured outreach to a defined population, is also the best destination for displaced claims and billing workers. It is a real, credentialed, trainable role that does not require a clinical degree.
• Alaska, the Nuka System of Care. American, Indigenous-owned, and the best primary care model in the country that nobody copies. Southcentral Foundation took over an IHS facility under the Indian Self-Determination and Education Assistance Act, then rebuilt care around customer-ownership (the people served own and govern the institution) and same-day access, with integrated behavioral health and large reductions in emergency and inpatient use. It works here, under American conditions, run by the community it serves.
• Netherlands: GP out-of-hours cooperatives. Regional cooperatives of family doctors covering nights and weekends, which is why the Dutch don't use emergency rooms as primary care. This is the direct fix for America's ER-as-front-door problem. A third of American physicians are in primary care against half or more in peer countries, so patients who cannot reach a GP use the emergency room instead. The cooperative model costs almost nothing compared with the emergency department it replaces.
• Netherlands: Buurtzorg. Nurse-led home care in self-managing neighborhood teams of about a dozen, with almost no management layer, better outcomes, higher staff satisfaction, and lower cost. Copied worldwide. Build the long-term care benefit on this model, small, self-directing clinical teams with minimal overhead, not on the American agency staffing model, which layers administrative management over fragmented, often poverty-wage care workers.

Quality — cheap, and America has almost none of it

• Sweden: national quality registries. About a hundred disease- and procedure-level registries recording every case nationally, with results published and every clinician able to see their own outcomes against their peers'. America's equivalent is fragmentary and mostly proprietary. Adopt wholesale; the national record makes them nearly free to run.
• UK: mandatory national clinical audit and the National Joint Registry. Same idea, procedure-specific, with public reporting.
• Denmark: no-blame incident reporting, which produces far more reported safety events than a punitive system does. Pairs with a no-fault compensation scheme for medical injury, in which patients are compensated without having to prove negligence and clinicians report errors without fear of lawsuit, a model described in the malpractice and liability provisions elsewhere in this plan.

Long-term care and aging

• Japan: Kaigo Hoken. Mandatory long-term care insurance, contributions from 40, needs assessed by a standard national instrument, and a care manager assigned to each person to assemble their package. (in, the long-term care and home care provisions described earlier in this plan)
• Denmark: stop building nursing homes. Denmark deliberately stopped building institutions and shifted to home care plus preventive home visits to everyone over 75, cheaper, and preferred by the people it serves. Adopt the preventive visit: it is a small, popular program that catches falls, isolation, and deterioration early. Catching those problems early keeps people out of the hospital, so costly hospitalizations fall.

Workforce

• Israel, mass absorption of foreign-trained physicians. Israel licensed tens of thousands of Soviet-trained doctors in the 1990s through an examination and bridging pathway, at speed and at scale. This is the working precedent for a foreign-trained physician pathway.
• Japan, regional bed quotas as workforce policy. Where the beds are permitted determines where the clinicians go. (in, the capacity section of the plan)
• Service obligations tied to publicly funded training. (in, the professions section of the plan)

Prevention and public health

• Finland, the North Karelia project. A community-wide campaign that cut cardiovascular mortality in the target region over several decades. It was community-organized rather than clinical, and it changed a food supply as well as what people were told. It is a community-organized, upstream intervention.
• Portugal, decriminalization plus treatment capacity. The relevant lesson is that Portugal paired decriminalization with an expansion of treatment. Directly applicable to addiction treatment build-out.
• UK, the soft drinks industry levy, which worked mainly by driving reformulation rather than by changing consumer behavior. A cheap, evidence-backed, unglamorous win.

Governance, and the best precedent of all

• Taiwan, how the system was built. Taiwan created its single-payer NHI in the mid-1990s, in a few years, after deliberately studying every other country's system first and taking the best pieces. It is the one major system that was designed rather than inherited, and it did what this plan proposes. Every other model on this list evolved over a century; Taiwan built one in a few years. It should be the international touchstone the way Hill-Burton is the domestic one.
• Estonia, X-Road and the patient-visible access log. (in, the record section of the plan)
• Maryland, all-payer hospital global budgets. (in, the payer section of the plan)

The five I would take first

Germany's AMNOG (a drug's price follows its proven benefit), Costa Rica's EBAIS teams (primary care teams assigned by neighborhood), Sweden's national quality registries (published outcomes for every case, high return, nearly free to run on the national record), Taiwan's floating point value (a fixed budget that cannot overrun), and the Netherlands' GP out-of-hours cooperatives (family doctors on call at night, so the ER stops being the front door). (cut)