In eating disorders like anorexia, where weight restoration is often an initial goal, it makes sense to consider whether the rate of weight gain can predict long-term treatment success. This seems intuitive: if someone makes early, consistent progress toward this important benchmark, aren’t they more likely to achieve lasting recovery than someone who struggles to gain weight? As our latest research reveals, the answer isn’t so black and white.

Our research team recently examined the link between early weight gain and anorexia treatment success using real-world data from over 200 patients with anorexia who received virtual family-based treatment (FBT). We wanted to test whether early weight gain alone can serve as a reliable benchmark for remission—and what actually works better for guiding treatment decisions.

What the study looked like

This study, which was published in Eating Disorders: The Journal of Treatment and Prevention, analyzed 233 patients with anorexia nervosa (AN) who received virtual outpatient FBT through Equip. We examined whether early weight gain—specifically, weight change by week eight—could accurately predict remission by week 20. Remission was defined using several clinical markers, such as normalized weight and symptom reduction.

To assess this, we used several different methods:

• ROC (receiver operating characteristic) analysis, a common statistical method for classifying outcomes based on thresholds.
• Logistic regression models, which allowed us to consider more variables (like each patient’s starting weight as a percent of their individualized target).

This side-by-side comparison helped us assess not just whether early weight gain works as a predictor—but how accurately and fairly it does so for diverse patients.

What we learned

At Equip, we recognize that recovery is rarely linear and never one-size-fits-all, and our findings further strengthened this belief, driving home just how individual the treatment and recovery journey can be. Below are some of our key takeaways.

1. Early weight gain alone doesn’t tell the full story

ROC analysis using early weight gain was only moderately accurate when it came to prediction. Even at the best-performing threshold—gaining 8.9 pounds by week 8—it still misclassified 36% of patients. That means more than a third of patients were incorrectly labeled as likely to succeed or fail based solely on this benchmark.

2. Including individualized context improves accuracy

When we added another variable—percent target weight at admission—to a logistic regression model, accuracy improved significantly. This model didn’t just offer a binary outcome (i.e., a “yes” or “no” on whether someone would recover)—it returned the probability of remission, which is far more useful for informing personalized care.

3. Rigid benchmarks don’t work for everyone

The core takeaway? Applying blanket benchmarks to determine who’s “on track” can be misleading and even harmful. Patients with different body types, growth patterns, or illness histories may gain weight at different rates, and that variability is normal.

4. Early progress matters—but not in isolation

While early weight gain does matter (it is associated with better outcomes), it’s only meaningful when viewed in context. Two patients may gain the same amount of weight in the first two months, but their clinical profiles and risk factors may suggest very different prognoses. Context is key.

How our findings will inform Equip treatment going forward

This study reinforces several core principles that guide Equip’s approach:

1. Individualized, data-informed care works best

Instead of relying on rigid milestones or numbers, we use individualized metrics—like target weights based on personal growth histories, not static BMI charts. Our goal is to tailor care to each patient’s unique journey, not force every person into the same mold.

Our providers also look at early progress within the full picture. They consider where a patient started, how they’re trending, and what other clinical signs suggest about engagement and medical stability.

2. Evidence-based care must be flexible and inclusive

This study is part of our broader mission to make sure that our research doesn’t just live in journals—it also shapes the care we provide. That means continuously evaluating what’s working, questioning outdated models, and creating pathways that serve all patients, not just those who fit neatly into a dataset.

3. We’re committed to expanding representation in research

While our sample was larger and more naturalistic than many clinical trials, we recognize that it wasn’t as diverse as it should be. Going forward, we’re working to increase representation across race, ethnicity, gender identity, and geography to ensure our care—and our data—truly reflects the communities we serve.

Recovery from anorexia isn’t a race to a number on a scale—and this study makes that clearer than ever. While early weight gain can be a helpful signal, it’s not a definitive answer. By using individualized, data-informed models that incorporate each patient’s unique context, we can better predict who will benefit from which type of care—and when they might need extra support.

At Equip, we’re committed to turning research into real-world impact. That means moving beyond one-size-fits-all benchmarks toward a more compassionate, nuanced understanding of progress, one that adapts to the patient, not the other way around.

Read the full paper here.