• Center on Health Equity and Access
  • Clinical
  • Health Care Cost
  • Health Care Delivery
  • Insurance
  • Policy
  • Technology
  • Value-Based Care

Differential Weight Loss Effects on Type 2 Diabetes Remission Among Adults

Publication
Article
Evidence-Based Diabetes ManagementMarch 2017
Volume 23
Issue SP4

An analysis of nationally representative survey-based data finds that 5.2% of adults with type 2 diabetes were in remission, without bariatric surgery, at the end of the second year.

INTRODUCTION: Little is known about the variation in the effect of nonsurgical weight loss among obese and nonobese individuals on the incidence of type 2 diabetes (T2D) remission.

METHODS: Using data from a nationally representative healthcare survey, we analyzed the differential effect of weight loss on the relationship between obesity and the incidence of T2D remission over the span of 1 year among 3755 adults. Anyone who reported having T2D in the first year, but not in the subsequent year, was considered to be in remission. Changes in a person’s weight were measured as change in the body mass index. Data gathered between 2009 and 2013 were analyzed in 2016.

RESULTS: The incidence of self-reported remission was 5.22% (P <.001). Among obese individuals (BMI≥30), those who experiences a 3% drop in weight, at minimum, were 2.1 percentage points more likely to report remission than those who lost less than 3% bodyweight (P <.05). Comparing all individuals who lost more than 3% of their weight with those who lost less than 3% of their weight, obese individuals were 3.7 percentage points more likely than nonobese individuals to report being in remission (P <.05). Furthermore, after accounting for demographic and clinical information, we found that T2D remission was negatively associated with the duration of a T2D diagnosis and diabetes medication type, and was positively associated with being African American.

CONCLUSIONS: T2D is reversible, and the relationship between obesity and T2D remission varies with weight loss. Introduction

Generally, type 2 diabetes (T2D) has been viewed as a chronic, progressive, and controllable, but irreversible, disease.1 Interventions that occur soon after diagnosis reduce the risk of macro- and microvascular disease and can slow disease progression.2 However, plasma glucose continues to increase regardless of the intensity of diabetes control or treatment type.3

Nevertheless, a number of clinical trials and case-control studies have, over time, reported a remission in T2D with bariatric weight loss surgery or with intensive lifestyle management.4-11 Recently, a retrospective study using a cohort of adults with T2D identified with administrative data from Kaiser Permanente, Northern California, suggested that T2D remissions do occur without bariatric surgery, but are rare.12 The study found that the cumulative incidences of partial and prolonged diabetes remission over a period of 7 years were 1.5% and 0.007%, respectively.

To our knowledge, the above mentioned study is the lone community-based remission incidence study. However, this study lacks generalizability to the US population. Therefore, we used the Medical Expenditure Survey (MEPS), a national survey of the US noninstitutionalized civilian population, to study the differential effect of weight loss among obese and nonobese adults in remission without bariatric surgery.

Methods

Data Sample. MEPS, an annual nationally representative survey, employs an overlapping panel design and includes a new panel each year. The survey collects health status, healthcare expenditures, and health insurance coverage information for each member of a sampled household for a period of 2 calendar years, through 5 rounds of interviews. In the Priority Condition (PC) section, each household member entering the survey for the first time is asked to indicate whether or not he or she has diabetes. In the third and fifth follow-up rounds, a person who reported having been told by a doctor or other healthcare professional that he/she had diabetes is asked to complete a self-administered Diabetes Care Survey (DCS).

At that point, if a patient says that the earlier response was an error or that he or she no longer has diabetes, the variable indicating diabetes diagnosis is reset to “No”. Otherwise, the respondent is asked to fill out the DCS survey in which he or she will be asked to indicate receipt of diabetes diagnosis. If a respondent reports being unaware of having diabetes, that person is not assigned a positive DCS weight. This weight adjusts for DCS nonresponse and standardizes to the number of persons with diabetes in the US civilian noninstitutionalized population.

The age of a diabetes diagnosis was not made available until the 2009 MEPS public use files. Therefore, this study includes individuals in panels 14 to 17 who reported having diabetes during the first-year PC survey section, as well as in the DCS, and who fully participated in both years of data collection. About 5% of these individuals had missing values for body mass index (BMI) and/or age when they received their diabetes diagnosis. We used a mulmultiple imputation method to assign the missing values of these 2 variables. Furthermore, the final analytic sample excluded those who had any of the following characteristics: (1) bariatric surgery or related surgical complications during the 2 years of their respective panels, (2) the absence of positive diabetic survey weight in the first year of their panel, (3) a BMI below 20 and/or age younger than 30 years when diagnosed with diabetes, and (4) use of an insulin-only medication regimen. A total of 3755 adults were in the analytical sample and 191 (5%) of them had an imputed BMI or age at which they were diagnosed for diabetes. Data gathered between 2009 and 2013 were analyzed in 2016.

Measures. Anyone who reported a diabetes diagnosis in both the PC section of the survey and in his or her first year of the DCS, but reported the absence of a diabetes condition (second year diabetes indicator variable = 2 [“No”]) in the subsequent year’s DCS, was considered to be in diabetes remission.

Individuals with a BMI less than 30 were categorized as nonobese. A change in an individual’s weight was measured in terms of a change in his/her BMI. A dichotomous indicator variable indicating a drop of more than 3% in BMI between the first and second year was constructed. The age reported in the latest round of the first year was used to categorize individuals into 3 age groups: 18 to 44 years, 45 to 64 years, and 65 years or older; 3 categorical variables indicating the years since initial T2D diagnosis were used. Next, the self-reported medication use in the DCS survey at the end of first year was used to construct 4 dummy variables to indicate if an individual was taking: (1) any medicine, (2) an oral agent only, (3) insulin only, or (4) a combination of insulin and oral agents. Anyone with a self-reported diagnosis of coronary heart disease, myocardial infarction, or stroke was designated as having cardiovascular disease. Other indicator variables included hypertension, gender, and race/ethnicity (Asian, African American, and other races).

Statistical Analysis. This study used Stata SE 14.1 (Stata Corporation, College Station, Texas) to predict T2D remission using a logistic regression model that accounted for the complex survey design. The DCS weights were used in the analysis. To assess model fit, the goodness-of-fit test accounting for survey design suggested by Archer and Lemeshow was conducted.13 Interaction effects and standard errors were computed as suggested by Norton, Wang, and Ai,14 but only after accounting for the complex survey design.

Results

A total of 3755 individuals from 4 panels (14 to 17) constituted our study sample. Among these individuals, 5.91% were in remission, 56.5% were obese, 45% were male, 33.7% saw more than a 3% drop in BMI in 1 year, 46.8% were 45 to 65 years old, 10.4% were given a T2D diagnosis within the last 2 years, and about 11% reported taking no T2D medication (Table 1). Those who reported being diagnosed for hypertension or cardiovascular conditions in their lifetime represented 77.6% and 34.1% of the sample, respectively.

Table 2 presents the results of the logistic regression analysis. The goodness-of-fit test indicated that the model was a good fit {F(9,180)=0.425 and P >F = .92}. Both nonobesity and weight loss (BMI loss) were positively associated with remission (P <.05). The relationship between the likelihood of remission and weight loss varied with obesity, in that nonobese individuals who lost weight were less likely to have remission (P <.015). Remission was also negatively associated with duration since the initial T2D diagnosis and use of diabetic medication at baseline. Being African American was positively associated with remission (P <.05).

Our results suggest that more than 5% of individuals reported remission in 1 year (see Table 3). Among obese individuals, those who underwent a more than 3% drop in their weight were 2.1 percentage points more likely to report remission than those who lost less than 3% of their weight (P <.05). Among those who lost less than 3% of their weight, obese individuals were 1.9 percentage points less likely to report remission than nonobese individuals (P <.05). Comparing the average predicted probability of remission for those who lost more than 3% of their weight with those who lost less than 3% of their weight, obese individuals were 3.7 percentage points more likely than nonobese individuals to report being in remission (P<.05). The likelihood of remission declines with the duration of T2D diagnosis. Compared with those who had T2D for 10 or more years, those who had T2D within the last 2 years were 6 percentage points more likely to be in remission (P <.001) and those having T2D for more than 2 years but less than 10 years were 1.7 percentage points more likely to be in remission (P = .05). Those taking no medication to treat T2D were more likely to be in remission than those taking oral agents, insulin only, or both insulin and oral agents by 19.3, 20.6, and 21.2 percentage points, respectively (P <.001). Those taking only oral agents were 2 percentage points more likely to be in remission than those taking both insulin and oral agents (P <.001). Compared with whites and Hispanics, African Americans were 2.4 percentage points more likely to be in remission (P <.05).

Discussion

In our analysis of nationally representative survey-based data, we found that 5.2% of adults with T2D reported being in remission (without bariatric surgery) at the end of the second year. To our knowledge, this is the first analysis providing evidence of remission among diabetic adults using data from a national survey representing the US noninstitutional population. Remission was highest among those not using any medication (23%) and occurred much less often among those using oral agents and insulin (1.3%). Our study further supports findings from earlier studies that indicate T2D is reversible in some cases.

Similar to 2 previous studies,9,12 the incidence of remission in our cohort was positively associated with fewer years since diagnosis, the absence of glucose-lowering medication intake, and being African American. In our study, we found no variation in remission with regard to an individual’s age. In addition, our study shows that the relationship between weight loss and diabetic remission differs between obese and nonobese individuals. Between those who lose more than 3% of weight and those who don’t, obese persons with T2D are more likely to report remission than nonobese adults with T2D.

Limitations

There are some limitations to our study. First, the presence and absence of T2D is based on self-reports, and therefore, the information is prone to errors. However, the findings of this study are in line with the findings of 2 previously published studies. In addition, both BMI and age values were missing in less than 5% of the sample. We used a multiple-imputation method to impute the missing BMI and age values. The results of our model did not change when individuals with missing BMI and age values were excluded from the analysis.

Author Information:

Virender Kumar, PhD, is with survey operations of Westat, Inc, Rockville, Maryland. William Encinosa, PhD, is with the Center for Delivery, Organization and Markets, Agency for Healthcare Research and Quality (AHRQ), Rockville, Maryland; and the McCourt School of Public Policy, Georgetown University. Hena Thakur is a third-year medical student at Boston University Medical School and a fellow with the National Institutes of Health, Rockville, Maryland. Kisha Thakur is an undergraduate in biology, University of Maryland, College Park, Maryland. The authors received no financial support from any institution. The views herein do not represent the views of AHRQ or HHS.

Corresponding Author:

William Encinosa, PhD.

Georgetown University

Old North Suite 100, 37th and O Streets, N.W.

Washington D.C. 20057

301-427-1437

wee4@georgetown.edu

References

1. Lim EL, Hollingsworth KG, Aribisala BS, Chen MJ, Mathers JC, Taylor R. Reversal of type 2 diabetes: normalization of beta cell function in association with decreased pancreas and liver triacylglycrol. Diabetologia. 2011;54(10):2506-2514. doi: 10.1007/s00125-011-2204-7.

2. Ramlo-Halsted BA, Edelman SV. The natural history of type 2 diabetes: practical points to consider in developing prevention and treatment strategies. Clinical Diabetes. 2000;18(2):80-84.

3. UK Prospective Diabetes Study Group. Intensive blood glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes. Lancet. 1998;352(9131):837-853.

4. Schauer PR, Bhatt DL, Kirwan JP, et al; STAMPEDE Investigators. Bariatric surgery versus intensive medical therapy for diabetes—3-year outcomes. N Engl J Med. 2014;370(21):2002-2013. doi:10.1056/NEJMoa1401329.

5. Arterburn De, Bogart A, Sherwood NE, et al. A multisite study of long-term remission and relapse of type 2 diabetes mellitus following gastric bypass. Obes Surg. 2013;23(1):93-102. doi: 10.1007/s11695-012-0802-1.

6. Mingrone G, Panunzi S, De Gaetano A, et al. Bariatric surgery versus conventional medical therapy for type 2 diabetes. N Engl J Med. 2012;366(17):1577-1585. doi: 10.1056/NEJMoa1200111.

7. Schauer PR, Kashyap SR, Wolski K, et al. Bariatric surgery versus intensive medical therapy in obese patients with diabetes. N Engl J Med. 2012;366(17):1567-1576. doi: 10.1056/NEJMoa1200225.

8. Pournara DJ, Osborne A, Hawkins SC, et al. Remission of type 2 diabetes after gastric bypass and banding: Mechanisms and 2 year outcomes. Ann Surg. 2010;252(6):966-971. doi: 10.1097/ SLA.0b013e3181efc49a.

9. Gregg EW, Chen H, Wagenknecht LE, et al; Look AHEAD Research Group. Association of an intensive lifestyle intervention with remission of type 2 diabetes. JAMA. 2012;308(23):2489-2496. doi:10.1001/jama.2012.67929.

10. Buchwald H, Avodor Y, Braunwald E, et al. Bariatric surgery: a systematic review and meta-analysis. JAMA. 2004;292(14):1724-1737. doi: 10.1001/jama.292.14.1724

11. Pories WJ, Caro JF, Flickinger EG, Meelheim HD, Swanson MS. The control of diabetes mellitus (NIDDM) in the morbidly obese with the Greenville Gastric Bypass. Ann Surg. 1987;206(3):316-323.

12. Karter AJ, Nundy S, Parker MM, Moffet HH, Huang ES. Incidence of remission in adults with type 2 diabetes: The Diabetes & Aging Study. Diabetes Care. 2014;37(12):3188-3195. doi: 10.2337/dc14-0874.

13. Archer KJ, Lemeshow S. Goodness-of-fit test for a logistic regression model fitted using survey sample data. Stata Journal. 2006;6(1):97-105.

14. Norton EC, Wang H, Chunrong Ai. Computing interaction effects and standard errors in logit and probit models. Stata Journal. 2004;4(2):154-167.

Related Videos
Ian Neeland, MD
Chase D. Hendrickson, MD, MPH
Steven Coca, MD, MS, Icahn School of Medicine, Mount Sinai
Matthew Crowley, MD, MHS, associate professor of medicine, Duke University School of Medicine.
Susan Spratt, MD, senior medical director, Duke Population Health Management Office, associate professor of medicine, division of Endocrinology, Metabolism, and Nutrition,
Stephen Nicholls, MD, Monash University and Victorian Heart Hospital
Amal Agarwal, DO, MBA
Dr Robert Groves
Dr Robert Groves
Related Content
© 2024 MJH Life Sciences
AJMC®
All rights reserved.