Weight management that scales without the cost spiral

GLP-1 medications have demonstrated that significant weight loss is biologically achievable. But real-world employer experience tells a more complicated story: pharmacy budgets are being consumed, adherence is poor, and discontinuation leads to weight regain — negating both clinical and economic benefits.

Sources: EBRI/BCBSA simulation model (2025); Blue Health Intelligence claims analysis; Prime Therapeutics persistence data; WTW employer pulse survey (2025); SURMOUNT-4 trial; BCBSA Minnesota school district case.

And the non-pharmacological alternatives have their own gaps

Key to Health combines a handheld breath acetone sensor with a mobile app that delivers a Mediterranean-style dietary framework — aligned with the 2026 U.S. Dietary Guidelines. When a member breathes into the device, they see a real-time measure of whether their body is burning stored fat. This biofeedback loop creates the behavioral reinforcement that sustains long-term change.

The program is entirely self-directed. There is no coaching staff, no physician oversight, and no pharmacotherapy. This means marginal cost per additional member approaches zero — making it feasible to offer to your entire eligible population, not just a high-risk subset.

Randomized controlled trial

Falkenhain et al. Obesity. 2021. Independent investigators at UBC. Funded by CIHR.

A pragmatic RCT comparing Key to Health against WW (an evidence-based active comparator, not placebo). Independently conducted, publicly funded, pre-registered.

Real-world evidence at scale

Submitted to Obesity and ObesityWeek Journal Symposium, 2026.

A pre-registered observational cohort study testing whether the RCT findings hold in real-world conditions — at 73× the enrollment, with 2-year follow-up. Intention-to-treat analysis with MMRM. No cherry-picking completers.

To our knowledge, this is the largest long-duration effectiveness dataset for a non-pharmacological, fully automated digital weight management program. Specific outcomes will be shared here upon peer-reviewed publication.

Mechanism evidence: why it works

Falkenhain et al. JMIR mHealth and uHealth. 2022.

Mediation analysis showing that biofeedback engagement drove dietary adherence, which drove weight loss. This isn't a black box — there's published evidence for why the program produces outcomes, giving confidence that results are reproducible across populations.

Before GLP-1s: structured step therapy

Require Key to Health as a first step before GLP-1 authorization. Members use the program for 2–3 months while our algorithm identifies who is responding to dietary behavior change alone — and who may benefit from pharmacotherapy. Members who respond never need the prescription. Members who don't have demonstrated medical necessity for escalation.

Your published evidence base — an independent RCT, 2-year real-world outcomes consistent with the trial, and FDA submission — gives your benefits team the clinical justification for a step-therapy requirement. This isn't a delay tactic. It's an evidence-based first-line option that happens to save significant pharmacy spend.

During & after GLP-1s: behavior skills for maintenance

GLP-1 medications change appetite, but they don't teach dietary skills. When members discontinue — and published data show most do within the first year — they haven't learned what to eat or how their choices affect their metabolism. Key to Health provides the dietary education and real-time metabolic feedback that supports members during GLP-1 therapy and helps them build the habits needed to maintain progress if they step down or discontinue.

Instead of GLP-1s: a standalone evidence-based option

For organizations that have excluded GLP-1 coverage for weight management — or are considering it due to cost — Key to Health offers a clinically studied alternative that can be offered to your entire eligible population. Fully automated, no coaching infrastructure, no ongoing pharmacy spend. Real-world outcomes at 2 years that met pre-specified endpoints in an intention-to-treat analysis of more than 11,000 adults.