Bariatric Surgery as the Founding Discipline of Trajectory Engineering

From Metabolic Intervention to a Planetary Infrastructure

Robin Blackstone, MD
Blackstone Health
IBC University of Oxford 2026 Congress
September 16, 2026

Editorial note: The talk title is preserved as delivered at Oxford. The framework formerly called “Trajectory Engineering” is now formally named Health Trajectory Engineering. The updated name is used below when referring to the framework in the present tense.

Transcript

I am a bariatric surgeon. For most of my career, I did one thing extraordinarily tangible: in about ninety minutes, under general anesthesia, I could perform an intervention that changed the course of a person’s health for years, sometimes decades.

But over time, that created a larger question for me.

If we can change the course of health so dramatically with a single intervention, why is the rest of the health system still organized around episodes, visits, procedures, admissions, and claims, rather than around the course of a person’s health over time?

That question led me beyond bariatric surgery. It led me to the idea of the health trajectory, to Health Trajectory Engineering, and ultimately to Health 4.0: a health system designed to understand, influence, and sustain a person’s health across the life course.

What I want to do today is show you how bariatric surgery became the place where that idea first became visible to me.

One intervention. A thirty-year curve.

Here is what a bariatric operation actually does. It is not simply a weight-loss procedure. It is a metabolic intervention that can bend a decades-long curve.

In the Swedish Obese Subjects study, adjusted mortality was about twenty-nine percent lower in the surgery group over long-term follow-up. In STAMPEDE, at five years, 29 percent after gastric bypass and 23 percent after sleeve gastrectomy achieved an HbA1c of six percent or less, compared with 5 percent with intensive medical therapy. Medication burden also fell.

The point does not require exaggeration. Blood pressure, sleep apnea, glycemic control, mobility, and cardiovascular risk can all change when an intervention is placed at the right point on the curve.

Bariatric surgery gave me the first clinical proof that trajectories are real, measurable, and alterable.

The health trajectory, defined

Let me say precisely what I mean by a trajectory, because this is the foundation for everything that follows. I published the definition this week as an open-access note.

A person’s health trajectory is the course of their health across the whole of life, treated as one continuous, individually attributed object of measurement. It is assembled from three interacting domains, organized from upstream to downstream: culture, environment, and biology. The person’s own account runs through all three as first-person data.

That ordering does not assign a fixed percentage of health to each domain. The domains interact continuously. It is causal and directional: culture helps shape institutions, policies, norms, and behavior; those forces create the environments in which people live; and those environments interact with biology across the life course.

Medicine traditionally enters the sequence largely at biology. Health Trajectory Engineering begins further upstream.

The comparison remains within the person: realized trajectory against expected trajectory. The difference is the engineered gain.

The Health Trajectory: A Definition, v1.0, published under CC BY 4.0.

The most trajectory-altering operation is paid as an episode

Here is the second reason bariatric surgery matters. One of medicine’s most trajectory-altering interventions is still paid for mainly as an episode: an admission, a procedure, a claim.

Then the patient goes home, and the work that determines whether the curve stays bent is much less consistently financed. That work includes nutrition, movement, medication adjustment, early trajectory signals, and years of follow-up.

Our specialty built an answer to part of that problem. MBSAQIP created standards, measurement, a national registry, and a quality-improvement discipline. In that sense, it is a continuity system born inside a Rescue specialty.

But a registry does not create a payer for longitudinal continuity. We had the intervention, the measurement, and the quality infrastructure. We still hit the financing wall.

The second proof—and the same wall

And now there is a second proof.

GLP-1-based therapies can alter weight, cardiovascular risk, and the subsequent course of disease at population scale. In SELECT, semaglutide reduced major cardiovascular events by about twenty percent in people with obesity or overweight and established cardiovascular disease without diabetes. In the 2024 KFF poll, 12 percent of U.S. adults said they had ever taken a GLP-1 drug.

Withdrawal trials show that weight commonly returns and cardiometabolic improvements move back toward baseline. Real-world data also show that stopping, restarting, and switching are common. Those studies do not establish that everyone needs one fixed weekly dose for life, and the major outcome trials did not test intermittent management.

That leads to a Health Trajectory Engineering question: Can incretin therapy be dynamically dosed against an individual’s measured trajectory rather than prescribed indefinitely at a fixed schedule?

Some people may need continuous treatment. Others may ultimately use lower doses, less frequent treatment, switching, or re-initiation when the trajectory begins to drift. Those strategies still need to be tested.

The financing system does not ask that longitudinal question. It pays, or stops paying, for the drug as an episode.

The question reform answered

So let me move from the operating room to the financing system.

About ninety-two percent of Americans have health insurance. The United States spent 5.3 trillion dollars on health in 2024.

Coverage matters enormously. But coverage and system design are different questions. For sixty years, American health reform has repeatedly asked who is in the pool. That question is morally and practically necessary. It is not sufficient.

Once most people are in the pool, the dominant cost problem lies inside the insured system, in what the pool purchases and when it purchases it.

Health 4.0 asks the second question: what are we buying with coverage?

What the insured buy

What do we buy? A small fraction of people account for a very large fraction of annual healthcare spending. In 2022, the highest-spending five percent of Americans accounted for nearly half of measured healthcare expenditures.

That statistic shows spending concentration. It does not show that every dollar was preventable or that every service was Rescue.

The larger point is that the benefit architecture is far more reliable at financing expensive established disease than it is at financing longitudinal trajectory management. Nutrition, coaching, continuous monitoring, social and environmental support, and years of follow-up remain fragmented across benefits, programs, employers, and out-of-pocket payment.

My bariatric patients taught me this directly. The operation can be authorized. The continuous support needed to preserve the altered trajectory is much harder to find.

The payer changes before the benefit arrives

Now look at what happens when a health intervention produces benefits over decades.

The cost is paid today. The benefit may not appear for ten or twenty years. But the payer often changes long before then.

An employer may pay for an intervention at age forty-five. The employee may leave three years later. Their insurer may change several times. At sixty-five, Medicare may become responsible for the complications that the earlier intervention was designed to prevent.

So the organization that pays for improving the trajectory is often not the organization that receives the later financial benefit.

That is the structural problem: health unfolds across a lifetime, but financing is divided into short periods of responsibility.

Who has a financial stake in how long we live?

Here is the part that changed how I thought about the financing problem.

There are already enormous financial institutions whose costs depend directly on how long people live and how long they remain healthy.

Take an annuity. If I buy an annuity and live five years longer than expected, the company has to pay me for five additional years.

Life insurance works in the opposite direction. If I live longer, the company pays the death benefit later.

Long-term-care insurance is different again. If I remain healthier and avoid years of disability or institutional care, its costs may fall.

The same improvement in a person’s health can therefore create a financial cost for one institution and a financial benefit for another. Today, those relationships largely sit in separate businesses.

Our question is whether some of the long-term financial value created by improving health can help pay for the interventions that improve health today. We do not yet know the answer. That is the actuarial problem we are studying.

Health 4.0: the operating architecture

So what is the proposed system?

We call it Health 4.0: a person-governed, trajectory-oriented health system designed to produce health across the life course.

The Human Principal retains decision authority. Doctor AI, a clinician, a payer, or a model may inform a decision. None of them owns it.

The trajectory, rather than the encounter, is the unit of continuity. Health Trajectory Engineering provides the closed learning loop: sense, interpret, deliberate, decide, act, measure, and learn. Pulse and Rescue remain coupled. Excellent Rescue remains essential, but it becomes the backstop rather than the organizing principle. Authority, technology, financing, culture, leadership, and governance align around the same purpose.

At the center is the Triangle of Trust: the Human Principal, an accountable human health ally, and Doctor AI as the longitudinal digital ally.

The purpose is larger than technology or cost. Health matters because it expands what a human being is able to do and become.

What Doctor AI adds

Doctor AI brings three things that the current system cannot reliably provide.

First, continuity: a longitudinal memory that persists across clinicians, insurers, locations, and years.

Second, precision: not biology alone, but biology interpreted together with culture, environment, goals, and the person’s trajectory.

Third, trust. Not trust as a promise, but trust created through governance, auditability, and independent verification.

Doctor AI is one component of Health 4.0. It is not the system.

And one boundary is essential: Doctor AI does not own the decision. Decision authority remains with the Human Principal.

The Health Trajectory Benefit

If health unfolds across a lifetime, then the benefit designed to support it has to follow the person across that lifetime.

That is the idea behind the Health Trajectory Benefit.

It would belong to the Human Principal and stay with that person as employers, insurers, clinicians, and circumstances change.

It would finance the longitudinal work that happens before Rescue: measurement, follow-up, prevention, treatment, adaptation, and continued observation of the trajectory.

The objective is not simply to pay for Rescue differently. It is to change the trajectory early enough that preventable Rescue is progressively reduced and, where possible, eliminated.

Excellent Rescue remains essential for trauma, unpredictable disease, and what cannot reasonably be prevented. It becomes the backstop, not the organizing principle.

The financing mechanism for doing this at scale remains to be modeled and tested. The architectural principle is simpler: the benefit must follow the person for long enough to match the time over which health changes.

Where would Health 4.0 begin?

If this is the proposed system, where would we begin?

First, coverage should be completed. People still need access to excellent Rescue while we build something better upstream.

Then I would begin testing the Health Trajectory Benefit with self-insured employers. They already pay directly for much of the healthcare used by their employees, which makes them a practical place to test whether longitudinal investment can change health trajectories and later costs.

Medicare is different, but equally important. It remains financially responsible during the years when many consequences of earlier trajectories become visible.

Ultimately, the Health Trajectory Benefit cannot depend on where someone works or which insurer they happen to have. The destination is a benefit that stays with every Human Principal across the life course.

The destination is universal. The implementation can be staged.

This is not a criticism of the people working inside the current system. Most are trying to do the best they can within the responsibilities and incentives they have been given. The problem is that no one is responsible for the whole trajectory.

A planetary infrastructure

Now we can widen the aperture.

The 2026 Global Burden of Disease analysis found that point estimates of the morbidity gap widened in 203 of 204 countries and territories from 1990 to 2023. We are getting better at extending life than at extending healthy life.

Health Trajectory Engineering observes, models, and intervenes on the propagation of health and disease through time. It works across cultural, environmental, and biological domains, at population scale and individual resolution.

The United States is a useful Region Zero because the mismatch is costly, visible, and measurable. But this is not only an American problem, and it is not only a healthcare problem. It concerns how societies organize themselves around health across the life course.

The outcome can be universal without the implementation being uniform. Culture, consent, financing, and trusted institutions differ. The data commons required for this work must be federated and reciprocal.

Health has failed by what it decided to buy

I came to this by way of a stapler and patients whose trajectories changed dramatically, then became difficult for the system to see because the system was organized around the next encounter rather than the life between encounters.

A country that cannot see people until they are already falling has not been failed by its clinicians. It has been failed by what that country decided to buy.

The product does not have to be an encounter, a claim, or an AI model. The product can be health across the life course, in service of what that person is able to do and become.

We can decide to buy something else.

Thank you.

References

• Sjöström L, et al. Effects of bariatric surgery on mortality in Swedish obese subjects. N Engl J Med. 2007;357:741–752.
• Schauer PR, et al. Bariatric surgery versus intensive medical therapy for diabetes, 5-year outcomes (STAMPEDE). N Engl J Med. 2017;376:641–651.
• Lincoff AM, et al. Semaglutide and cardiovascular outcomes in obesity without diabetes (SELECT). N Engl J Med. 2023;389:2221–2232.
• West J, et al. Weight regain after cessation of medication for weight management: systematic review and meta-analysis. BMJ. 2026.
• Gasoyan H, et al. Obesity treatments and weight changes in clinical practice after discontinuation of semaglutide or tirzepatide. Diabetes Obes Metab. 2026.
• Boston University School of Public Health. GLP-1 discontinuation and reinitiation in adults with type 2 diabetes. ENDO 2026.
• KFF Health Tracking Poll, May 2024.
• U.S. Census Bureau. Health Insurance Coverage in the United States: 2024 (P60-288).
• CMS Office of the Actuary. National Health Expenditures 2024 Highlights and Projections 2024–2033.
• AHRQ MEPS. Concentration of Healthcare Expenditures, 2018–2022.
• CDC. Chronic Disease Facts, 2025.
• Whaley CM, et al. Prices Paid to Hospitals by Private Health Plans, Round 5.1. RAND; 2024.
• U.S. Bureau of Labor Statistics. Employee Tenure in 2024.
• NAIC. U.S. Life and A&H Insurance Industry: 2024 Annual Results.
• IMF. The Financial Impact of Longevity Risk. Global Financial Stability Report, April 2012, chapter 4.
• Garmany A, Terzic A. Global healthspan-lifespan gaps among 183 WHO member states. JAMA Netw Open. 2024;7(12):e2450241.
• Hay SI, et al. Global, regional, and national trends in the morbidity gap, 1990–2023 (GBD 2023). Lancet Public Health. 2026.
• Blackstone RP. Doctor AI: Reimagining Healthcare, Rebuilding Trust, Delivering Health 4.0. Blackstone Press; 2026.
• Blackstone RLP. Trajectory Engineering: A Prospectus; Health 4.0 and the End of Zero-Sum Healthcare. 2026.

In preparation: Blackstone, Pronk, Arena, Woodard. Health 4.0: A Person-Governed Architecture for the Continuous Production of Health. Blackstone, Minnaar, Sammut, Pronk. The Coverage Illusion. Blackstone. The Longevity-Short Balance Sheet.