Current status of anti-obesity medications and performance, an EHR based survey

Over the past two decades, the Food and Drug Administration (FDA) has significantly increased the approval of anti-obesity medications (AOMs) for obesity management. Both FDA-approved AOMs (F-AOMs) and off-label AOMs (O-AOMs) have gained popularity and demonstrated promising results in randomized clinical trials (RCTs). However, their effectiveness in real-world settings remains less understood. In this study, we evaluated population-level responses to AOMs and individual variability under real-world conditions, leveraging electronic health records (EHRs) as the primary data source for a comprehensive analysis of obesity relevant metrics.

Methods

EHRs of patients with obesity or overweight diagnosis were retrieved from the University of Texas Physician (UT-Physician) and EPIC COSMOS database. F-AOMs and O-AOMs were analyzed for their effects on obesity relevant metrics including body weight, body mass index (BMI), blood pressure (BP), high-density lipoprotein (HDL), low-density lipoprotein (LDL), HbA1c, and triglyceride levels.

Results

From the UT-Physician (U) and COSMOS (C) datasets, we identified 444K (2005–2023) and 30M (2018–2023) patients as obese or overweight, respectively. Of these, 72,089 (6.2%) and 10M (∼30%) had exposure to AOMs. During the study period, [U:37%; C:48%] of patients underwent more than one treatment session. The median exposure time to F-AOMs are [U:2.9; C:6.1] months. The median exposure time to O-AOMs are [U:2.9; C:14] months. For both cohorts, F-AOMs generally showed superior weight loss effects than O-AOMs. Tirzepatide and semaglutide were the most effective F-AOMs. Tirzepatide achieved 5% and 10% weight loss respectively in [U:65%; C:63%] and [U:33%; C:42%] of patients on long-term [U:76 wks; C:79 wks] exposure. Semaglutide achieved 5% and 10% weight loss respectively in [U:37%; C:45%] and [U:16%; C:25%] of patients on long-term [U:94 wks; C:88 wks] exposure. Both medications also benefited HDL, LDL, triglyceride, and HbA1c levels. Among O-AOMs, metformin demonstrated the most effective long-term [U:143 wks; C:166 wks] results, with [U:27%; C:32%] and [U:9%; C:14%] of patients achieving 5% and 10% weight loss, respectively. Significant individual variability in response was observed across all AOMs, irrespective of diagnosis type or exposure duration. Weight gain was noted in a subset of patients, ranging from [U:15%; C:18%] for tirzepatide (long-term) to [U:59%; C:62%] for lisdexamfetamine (long-term). Significant weight regain was consistently observed in both cohorts after discontinuing phentermine series, zonisamide, topiramate, and lisdexamfetamine.

Conclusions

Remarkable differences were observed between real-world evidence from EHR data and RCT reports on AOM effectiveness, highlighting the challenges of applying RCT results to diverse populations in routine practice. The wide variation in individual responses to both FDA-approved and off-label AOMs underscores the importance of personalizing obesity treatment to improve effectiveness, support long-term adherence, and promote sustainable care strategies.