Home / Conditions / Obesity / How Does Weight Fluctuation Affect Type 2 Diabetes?

How Does Weight Fluctuation Affect Type 2 Diabetes?

Feb 16, 2019
Editor: Steve Freed, R.PH., CDE

Author: Annahita Forghan, Pharm.D. Candidate 2019, LECOM College of Pharmacy

Major changes in weight, including weight loss, can affect diagnosis risk.

In type 2 diabetes, it is important to counsel patients on managing their body weight due to evidence of the negative impact and risk that excess weight (especially obesity) has on the disease. Research has shown that the American population’s average weight has been rising. According to data from the National Health and Nutrition Examination Survey, two of three American adults have either excess weight or obesity… about 30% of these adults have diabetes and about 85% of patients with type 2 diabetes mellitus are  has excess weight. But research has not been done much on the impact of oscillations in body weight on type 2 diabetes. Is a great fluctuation in weight harmful as well, even if it involves bouts of weight loss?

In this study, each year, data was taken from a public dataset from the Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial: “a multicenter factorial randomized controlled trial that compared intensive blood pressure, glycemic and lipid treatment with standard care in patients with diabetes mellitus.” Baseline weight, change in weight, and body weight variability (BWV) were analyzed to see if they were linked to type 2 diabetes. According to the study, change in weight was defined as “baseline weight minus exit /final weight (kg).” Body weight variability was defined as “average successive variability in weight (average absolute difference between successive values) during the trial.” There were 10,251 people who participated. Patients with hemoglobin A1C ≥7.5%, type 2 diabetes mellitus, ages 40 to 79 with CAD or 55 to 79 years with anatomical evidence of significant atherosclerosis, albuminuria, left ventricular hypertrophy or ≥2 cardiovascular risk factors (dyslipidemia, hypertension, current smoking, and obesity) were included. Patients unwilling to perform home glucose monitoring or insulin injections, with frequent or recent serious hypoglycemic events, with BMI > 45, serum creatinine > 1.5 mg/dL and serious illness were excluded.

Participants who gained the most weight (gained > 4.1 Kg), participants who gained mild weight (gained between zero and 4.1kg), participants with mild weight loss (0-3.8kg), and participants who lost the most weight (> 3.8kg) were divided into 4 quartiles of change in weight (Quartile 1 [Q1] being the ones who gained the most, and Q4 being the ones who lost the most) and then compared. The diabetes outcomes risks were also evaluated between the participants with normal weight but highest quartile of body weight variability, and participants with obesity at baseline but who had lowest quartile of body weight variability. SAS software 9.2 (SAS Institute, Cary NC) and excel spreadsheet were used to assess the data.

There were 911 (8.9%) of participants who were of normal weight (BMI [body mass index]> 25), 2,985 (29.1%) had excess weight (BMI 25 to 30), and 6,355 (62%- more than half) had obesity (BMI >30). The study used a Cox proportional hazards model to investigate relationships between baseline BMI, change in weight, and body weight variability. The primary outcome was defined as “nonfatal MI or nonfatal stroke or CV,” according to the study, and this composed 10.2% of the participants after about 3.5 years. In addition to the primary outcome, 60.7% of the patients had a microvascular complication (nephropathy, neuropathy or retinopathy), 4.3% had heart failure, and 7% died. There was a statistically significant correlation found between body weight variability and the primary outcome, heart failure, death, and microvascular events in their subjects. “Increasing quartiles of body weight variability was associated with increasing number of CHF and microvascular events. However, for the primary outcome and total mortality, those with the least (Q1) and most (Q4) body weight variability appear to have high risk albeit the percentage of events were significantly higher in those with the highest body weight variability (Q4)… The highest percentage of total deaths appears to be concentrated in those with lowest (Q1) and highest (Q4) quartiles of body weight variability within each weight category.”1

The subjects who died had gained the most weight, but also, the body weight variability itself significantly demonstrated negative outcomes aside from CVD risk factors and BMI. Therefore, there is great risk linked with weight fluctuations in people with type 2 diabetes mellitus.

Practice Pearls:

  • It is important for patients to lose excess weight to prevent type 2 diabetes risk and complications, but body weight fluctuations come with risks as well, even if weight has been lost..
  • This study’s 10,251 participants were divided into 4 quartiles of weight variation, with Quartile 1 being the patients who gained the most weight, and Quartile 4 being the patients who lost the most weight. 10.2% obtained the primary outcome (nonfatal MI or nonfatal stroke or CV), 60.7% of the patients had a microvascular complication (nephropathy, neuropathy or retinopathy), 4.3% had heart failure, and 7% died.
  • It was shown that most of the deaths and primary outcomes occurred within the lowest (Q1) and highest (Q4) quartiles of body weight variability.


Bertoni, Alain G.; Hsu, Fang-Chi; Yeboah, Joseph; Yeboah, Phyllis. “Body Mass Index, Change in Weight, Body Weight Variability and Outcomes in Type 2 Diabetes Mellitus (from the ACCORD Trial).” American Journal of Cardiology. 15 February 2019. https://www.ajconline.org/article/S0002-9149(18)32097-6/fulltext. 5 February 2019.

Annahita Forghan, Pharm.D. Candidate 2019, LECOM College of Pharmacy