Millions of Americans taking GLP-1 receptor agonist drugs such as Ozempic are shedding pounds, but their daily physical activity appears to be dropping sharply at the same time. An analysis of wearable device data from the National Institutes of Health’s All of Us Research Program found that average daily steps fell from 5,047 to 4,487 and moderate-to-vigorous physical activity declined from 28 minutes per day to 22. The findings, presented in materials from The Endocrine Society, raise a pointed question: if the drugs suppress appetite and reduce body weight, are patients quietly losing the movement habits that protect muscle mass and long-term metabolic health?
Why a 560-step daily drop matters for GLP-1 drug users
The numbers look modest in isolation. A decline of roughly 560 steps per day and six fewer minutes of moderate-to-vigorous physical activity (MVPA) might not alarm someone watching the scale trend downward. Yet exercise science has long established that sustaining activity during weight loss helps preserve lean muscle, maintain bone density, and support cardiovascular fitness. Losing weight while becoming more sedentary can trade one health risk for another, replacing excess fat with accelerated muscle loss and reduced functional capacity.
GLP-1 drugs work by mimicking a gut hormone that slows gastric emptying and dampens hunger signals. Patients eat less, feel full sooner, and often report lower motivation to move. Materials summarizing the All of Us analysis for The Endocrine Society confirm that daily steps decreased from 5,047 to 4,487 among participants taking these medications. That 11 percent step reduction, paired with a 21 percent drop in MVPA minutes, suggests the drugs may be reshaping daily behavior in ways that clinicians and patients are not routinely tracking or discussing in the exam room.
For the roughly 15 million adults in the United States estimated to have received GLP-1 prescriptions in recent years, the practical takeaway is direct. Weight loss alone does not guarantee improved fitness. If patients and their doctors are not monitoring activity levels alongside body weight, the net health benefit of these drugs could be smaller than headline weight-loss numbers imply. A shrinking waistline may coexist with weaker muscles, lower stamina, and reduced resilience to illness or injury.
Fitbit records and electronic health data behind the findings
The study drew on a distinctive data source. The All of Us Research Program, run by the NIH, collects electronic health records, survey responses, biospecimens, and wearable device data from a diverse cohort of participants across the country. Researchers linked prescription records for GLP-1 drugs to Fitbit-derived activity metrics, creating a dataset that tracks real-world movement patterns before and after patients started these medications. That linkage allows investigators to observe behavior changes without relying on self-reported exercise logs, which are prone to recall bias.
A peer-reviewed methods paper in the National Library of Medicine’s archive details how All of Us processes Fitbit data, including issues such as date shifting and missing records. Fitbit’s proprietary algorithms determine how raw accelerometer signals translate into step counts and activity classifications. Different device models and firmware versions can produce slightly different readings for identical movements. The methods paper also describes valid-day rules that researchers apply to filter out days with insufficient wear time, which shapes how averages are calculated and how many days of data each participant contributes.
These technical details matter because they affect confidence in the reported step and MVPA declines. The All of Us wearables infrastructure, described in companion documentation accessible through the NIH’s central database, acknowledges that device heterogeneity and algorithmic opacity are known limitations. The gap between raw sensor output and a clean activity metric leaves room for measurement noise that could inflate or obscure real behavioral changes. Even so, the direction and magnitude of the shift-hundreds of steps and several MVPA minutes per day-are large enough that random error alone is unlikely to explain them.
Another strength of the All of Us approach is its longitudinal structure. Participants often contribute months of Fitbit data before and after a new prescription, giving researchers a built-in baseline. That design helps distinguish genuine behavior changes associated with treatment from seasonal fluctuations in activity, such as winter slowdowns or summer surges. However, the current analysis, as described in The Endocrine Society materials, has not yet been fully published in a peer-reviewed journal, limiting outside scrutiny of its statistical methods.
Is it the drug, the weight loss, or both?
One hypothesis worth testing is whether the observed drop in movement is driven by the drug itself or simply by the act of losing weight. People who shed significant pounds through diet alone sometimes reduce their activity as their caloric intake falls and energy levels shift. Fatigue, lightheadedness, or a narrowed focus on food restriction can all nudge individuals toward more sedentary routines, even as they celebrate lower numbers on the scale.
If matched cohorts of individuals who lost similar amounts of weight without GLP-1 drugs showed comparable step declines in the same All of Us Fitbit records, the activity reduction would look less like a medication-specific side effect and more like a general consequence of caloric deficit. Conversely, if GLP-1 users reduced their movement far more than non-users with similar weight loss, that would support the idea that appetite suppression and possible nausea or gastrointestinal side effects are uniquely discouraging physical activity.
The study materials released through The Endocrine Society do not include regression outputs linking the magnitude of weight lost to the size of the step or MVPA decline. No published tables break the results down by specific GLP-1 agent, such as semaglutide versus tirzepatide, or by duration of drug use. Without those stratifications, it is difficult to determine whether someone who lost 30 pounds on a weekly injection moved less than someone who lost 10, or whether the activity drop appeared early in treatment and then stabilized or reversed as patients adapted.
The absence of a non-pharmacologic comparison group is the analysis’s most significant gap. Researchers working within the All of Us dataset could, in principle, identify participants who lost weight through behavioral interventions, bariatric surgery, or other non-GLP-1 medications and who wore Fitbits during the same period. That comparison would clarify whether GLP-1 drugs carry a unique sedentary signal or whether any substantial weight loss tends to come with reduced daily movement, regardless of the pathway.
Clinical implications and what patients can do now
Even with unanswered questions, the emerging evidence has immediate relevance for clinicians prescribing GLP-1 drugs and for patients using them. Providers often emphasize dietary changes and side-effect monitoring when initiating these medications, but structured guidance on physical activity can be less explicit. The All of Us findings suggest that counseling should include a clear plan to maintain or increase movement, not merely a reminder to “stay active.”
Simple strategies can help counter the drift toward sedentariness. Clinicians might encourage patients to track steps or active minutes with their own wearables, set individualized daily targets, and schedule brief check-ins to review both weight and activity trends. Resistance training two or three times per week can help preserve muscle mass during rapid weight loss, while brisk walking, cycling, or swimming can maintain cardiovascular fitness. For patients experiencing nausea or fatigue on GLP-1 therapy, adjusting the timing or intensity of exercise-shorter sessions, lower impact activities, or movement breaks spread throughout the day-may make activity more sustainable.
For patients, the key message is that the scale is only one metric of progress. If pants sizes are dropping but stair climbing feels harder, or if daily step counts fall month after month, it may be time to revisit the treatment plan. Discussing these patterns openly with a healthcare provider can lead to dose adjustments, additional support for exercise, or referrals to physical therapy or supervised programs that emphasize safe, gradual conditioning.
As GLP-1 drugs move from niche obesity treatments into mainstream chronic disease management, understanding their full behavioral footprint will be critical. The All of Us analysis offers an early warning that weight loss achieved through powerful pharmacology may come with quieter shifts in how people move through their days. Capturing and addressing those shifts could determine whether the current wave of prescriptions translates into the durable, whole-body health gains that patients and clinicians expect.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
