OR07-2 Patterns of Weight Loss, Maintenance, and Gain in Patients with Obesity

Program: Abstracts - Orals, Poster Previews, and Posters
Session: OR07-Predictors of Weight Gain and Disease
Friday, April 1, 2016: 11:45 AM-1:15 PM
Presentation Start Time: 12:00 PM
Room 160 (BCEC)
Joanna Huang1, Sarah Buchs1, Maral DerSarkissian*2, Rachel Bhak3, Francis Vekeman4, Rahul Ganguly1 and Mei Sheng Duh3
1Novo Nordisk, Inc., Plainsboro, NJ, 2Analysis Group, Inc., Los Angeles, CA, 3Analysis Group, Inc., Boston, MA, 4Groupe d’Analyse, Ltée, Montreal, QC, Canada
Background: Achieving and maintaining weight loss (WL) has proven to be difficult as many patients regain weight after an initial WL.  To understand patterns of weight change among patients with obesity, we conducted a retrospective longitudinal study of patients with obesity using the General Electric Centricity electronic medical record database.  Identifying patterns of weight change is critical for tailoring weight management strategies to the needs of targeted patient groups.   

Methods: The study sample included patients age ≥18 years old with BMI ≥ 30 kg/m2 (first observed BMI measurement was defined as index BMI) who had no medical conditions associated with unintentional WL and had ≥4 BMI measurements per year for ≥5 years.  Patients were categorized into 4 weight groups based on their amount of weight change during a WL period of two quarters following the index BMI: stable weight= within <5% of index BMI; modest WL= ≥5 to <10% of index BMI lost; moderate WL= ≥10 to <15% of index BMI lost; and high WL= ≥15% of index BMI lost.  Patterns of weight change were also assessed following the WL period for 8 consecutive quarters (i.e., the weight maintenance [WM] period).  Subgroup analyses were conducted among patients with class II obesity (index BMI≥35), pre-diabetes, and type 2 diabetes (based on diagnoses and lab values).

Results: Of 194,490 patients, 177,743 patients were included in the main analysis as they were classified in one of the 4 weight groups of interest during the WL period: 151,236 (85.1%) patients were in the stable weight group, 16,559 (9.3%) were in the modest WL group, 4,017 (2.3%) were in the moderate WL group, and 5,931 (3.3%) were in the high WL group.  Though 7.2% of the modest WL group continued to lose weight in the first quarter of the WM period, this dropped to 2.0% by the eighth quarter; in the moderate WL group these proportions were 13.3% and 4.1%; and 18.8% and 11.1% for the high WL group.  A decreasing trend in the overall proportion of patients who maintained their weight throughout the WM period was also observed.  In the modest, moderate, and high WL groups 40.0%, 35.9%, and 18.6%, of patients, respectively, regained ≥50% of lost weight during the WM period.  Cyclers were defined as patients who did not consistently lose, maintain, or gain weight in each quarter of the WM period.  The high WL group had the lowest proportion of cyclers with 58.3%, while 71.5% of the modest WL group and 74.1% of the moderate WL group were cyclers.  This trend was similar in the subgroups analyzed, though among patients with class II obesity those with high initial WL were more likely to continue to lose weight during the WM period compared to the overall sample and other subgroups.      

Conclusion: Weight cycling and regain was commonly observed among patients.  Patients who lost more weight during the WL period (overall and in subgroups) were more likely to keep the weight off and continue losing weight.

Disclosure: JH: Employee, Novo Nordisk, Employee, Novo Nordisk. SB: Employee, Novo Nordisk. MD: Researcher, Novo Nordisk. RB: Researcher, Novo Nordisk. FV: Researcher, Novo Nordisk. RG: Employee, Novo Nordisk, Employee, Novo Nordisk. MSD: Researcher, Novo Nordisk.

*Please take note of The Endocrine Society's News Embargo Policy at https://www.endocrine.org/news-room/endo-annual-meeting/pr-resources-for-endo

Sources of Research Support: Supported by Novo Nordisk, Inc.