For most obesity cases, the path to a healthy weight runs through behavior. Our patented analytics read that behavior and forecast where each patient is headed. The next step in the platform extends the view — adding biology to behavior, so clinicians can distinguish what patients do from what their bodies are doing back.
The first principle
A weight reading is a snapshot of a body at one moment, shaped by hydration, glycogen, hormones, and the previous day's salt. Treat the snapshot as truth and you spend the program reacting to biological noise.
What predicts a patient's trajectory is not their current weight. It is what they are doing — how often they engage, how they respond to a difficult day, whether they keep showing up. Behavior is the signal under the noise. Reading it is the difference between a clinic that reacts to dropouts and a clinic that prevents them.
This is the foundation Ew2health was built on, and it is the principle the rest of the platform extends from.
Predictive Behavioral Analytics
The core technology of Ew2health is a machine learning model that learns each patient's natural pattern, separates behavior from biological noise, and forecasts trajectory.
Daily weight fluctuates by 0.5–2 kg from hydration, glycogen, hormones, and stress. Our model spends the first two weeks learning each patient's unique fluctuation signature — the shape of their natural noise — before it begins forecasting.
Once the baseline is learned, the model continuously separates biological variation from genuine behavioral change. A 1.2 kg jump after a salty dinner is recognized for what it is. A pattern shift over five days is recognized for what it is, too.
From there, the model projects each patient's weight trajectory up to fifteen days ahead. Clinicians do not review charts — they review who is at risk today, while there is still time to act.
The science
Three established findings shape how Ew2health is designed. None of them are ours. All of them inform what patients see, what clinicians see, and what the model looks for.
Pillar 01
People avoid monitoring information that feels threatening or judgmental. A patient who senses they had a difficult week often stops weighing in altogether — not because they have given up, but because the scale has become a verdict. Sinque is designed to remove that verdict. The patient sees a trajectory, not a number, and the threat dissolves.
Chang, Webb & Benn (2017)
Pillar 02
Behavior change happens in stages — precontemplation, contemplation, preparation, action, and maintenance — and the right intervention depends on which stage the patient is in. A patient in maintenance does not need a motivation nudge; they need stability cues. A patient drifting from action back toward contemplation needs early support, not late correction. Reading behavior tells the clinical team which stage each patient is actually in, regardless of what the chart suggests.
Prochaska & DiClemente (1983)
Pillar 03
Regular self-weighing is one of the strongest predictors of sustained weight management — when it does not become a source of shame. The clinical evidence is consistent: interventions that include frequent weight monitoring outperform those that don't, sometimes by a wide margin. The challenge has always been keeping patients engaged with the scale long enough to benefit from the data. That is the engagement problem Sinque was designed around.
Michie et al. (2009); VanWormer et al. (2009)
The next layer
GLP-1 medication has shown the world that willpower is not the whole story. Biology pushes back. The next evolution of Ew2health adds two biological signals to the behavioral model — so when a patient is doing everything right and the weight still is not moving, the platform surfaces what is happening underneath.
Glucose variability
Continuous glucose monitoring data will reveal whether a patient's metabolism is responding to dietary changes the way it should — or whether something deeper is interfering.
Cardiorespiratory fitness
VO₂max captures cardiorespiratory fitness as a downstream marker of overall metabolic health. Combined with weight trends and glucose variability, it gives clinicians a third dimension to triangulate what is actually happening in the patient's body.
Together with weight trends, these form the 3D biomarker model — not a replacement for behavioral monitoring, but a deeper view alongside it.
Validation
More than five hundred thousand weight measurements analyzed across multiple markets, real-world clinical deployments, and an active validation program with one of the most respected institutions in metabolic health.
In 2026, Sinque participated in the Mayo Clinic Platform Data Accelerate program — validating our analytics against tens of thousands of GLP-1 treatment records and contributing to the platform's data backbone.
Active in medical weight-loss clinics across the Netherlands, Brazil, and expanding to the United States. The platform is generating revenue and clinical signal in real practices, not lab conditions.
A 12-month and 6-month pilot at the Petrobras medical nutrition clinic produced documented improvements in retention and clinical outcomes compared to standard care. The pilot data informs current platform design.
What it means for your clinic
We built an interactive impact calculator that lets clinic owners walk through their own situation across four short chapters — the revenue they are losing today, the market pressure compressing margins, the recurring revenue model with Sinque, and finally the investment in honest detail. It takes about three minutes. There is no login and no contact form to get to the results.
If you would prefer to walk through the platform with us — and have your questions answered by people who built it — a 30-minute demo is the next step.
