Results Summary
What was the research about?
Some adults and youth with severe obesity can’t lose enough weight with diet and exercise to improve their health. Different types of surgery may help people lose weight and improve problems related to obesity like diabetes. But surgery has risks, including needing more surgery to reverse or fix the original surgery.
In this study, the research team compared three types of weight-loss surgery:
- In Roux-en-Y gastric bypass, or RYGB, a surgeon uses part of the stomach to create a pouch that holds a small amount of food. The surgeon attaches this pouch to the small intestine. With this bypass, the body absorbs fewer calories.
- In adjustable gastric banding, or AGB, a surgeon places a band around the upper part of the stomach. This band limits how much a person can eat.
- In sleeve gastrectomy, or SG, a surgeon removes a large part of the stomach, which also limits the amount of food a person can eat.
The research team looked at patients’ health records to see how surgery affected
- Weight loss in 46,510 adults and in 544 youth ages 12–19
- Effects on diabetes in 10,019 adults with type 2 diabetes
- The need for more surgery in 34,714 adults
What were the results?
Weight loss in adults. With all three types of surgery, patients lost weight and then regained some weight over five years. At five years, adult patients who had RYGB lost 26 percent of their total weight compared with before surgery. Patients who had SG lost 19 percent and patients who had AGB lost 12 percent.
Diabetes. Diabetes went away at some point in the five years after surgery for 86 percent of patients with diabetes who had RYGB, 84 percent of patients who had SG, and 65 percent of patients who had AGB.
Need for more surgery. Adults who had SG had a 9 percent chance of needing another surgery in five years compared with 12 percent of people who had RYGB and 41 percent who had AGB.
Weight loss in youth. One year after surgery, youth who had RYGB lost an average of 31 percent of their body mass index, or BMI. Youth who had SG lost an average of 28 percent of BMI, and youth who had AGB lost an average of 10 percent.
What did the research team do?
The research team looked at health data for patients who had a BMI of at least 35 and had surgery between 2005 and 2015. Patient data came from 41 health systems around the country that were part of PCORnet®.
Patients, advocacy groups, surgeons, and health insurers helped plan the study.
What were the limits of the study?
Patients and their doctors chose what type of surgery to have. Results may have differed if the research team had assigned patients by chance to the type of surgery. The study didn’t include outpatient surgery centers, which may do more AGB surgeries.
Future research could track patients’ weight regularly.
How can people use the results?
Patients with obesity and their doctors can use these results when considering surgery for weight loss.
| How this project fits under PCORI’s Research Priorities The PCORnet® Study reported in this results summary was conducted using PCORnet®, the National Patient-Centered Clinical Research Network. PCORnet® is intended to improve the nation’s capacity to conduct health research, particularly comparative effectiveness research (CER), efficiently by creating a large, highly representative network for conducting clinical outcomes research. PCORnet® has been developed with funding from the Patient-Centered Outcomes Research Institute® (PCORI®). |
Professional Abstract
Objective
To compare the safety and effectiveness of Roux-en-Y gastric bypass (RYGB), sleeve gastrectomy (SG), and adjustable gastric banding (AGB) in increasing weight loss and improving diabetes outcomes
Study Design
| Design Elements | Description |
|---|---|
| Design | Observational: cohort study |
| Population | Adult and adolescent patients with baseline BMI ≥35 who underwent bariatric surgery between January 1, 2005, and September 30, 2015 |
| Interventions/ Comparators |
|
| Outcomes |
|
| Timeframe | 5-year follow-up for primary outcomes |
This retrospective observational cohort study used data from 41 health systems in 11 PCORnet® Clinical Data Research Networks to identify adolescents and adults who had a preoperative body mass index (BMI) of 35 or greater and underwent primary bariatric surgery between 2005 and 2015. Researchers studied four cohorts:
- Adult weight loss cohort included 46,510 patients who had BMI available after surgery.
- Adult diabetes cohort included 10,019 patients with type 2 diabetes who had HbA1c lab values and medication data available before and after surgery.
- Adult safety cohort included 34,714 patients. Researchers linked electronic health records to claims data to measure additional operations or interventions.
- Adolescent weight loss cohort included 544 patients ages 12–19 at the time of surgery who had BMI available after surgery.
Adult cohorts were ages 20–79 at the time of surgery. Researchers examined outcomes for up to five years after surgery.
Patients, advocacy groups, surgeons, bariatric medicine specialists, and health insurers helped plan the study.
Results
Adult weight loss cohort. Patients in all treatment groups lost weight after surgery and regained some weight over five years. Compared with baseline, average weight loss was 26% for RYGB, 19% for SG, and 12% for AGB. Total weight loss was higher for RYGB than SG and AGB (both p<0.01) and higher for SG than AGB (p<0.001).
Adult diabetes cohort. At five years, 86% of patients who had RYGB had experienced diabetes remission (HbA1c<6.5%) at some point, compared with 84% for SG and 65% for AGB. Incidence of remission was higher for RYGB than for SG (p=0.007) or AGB (p<0.0001).
Adult safety cohort. Cumulative probability of another operation or intervention over five years was 9% for SG, 12% for RYGB, and 41% for AGB.
Adolescent weight loss cohort. At one year, adolescents undergoing RYGB lost an average of 31% of baseline BMI compared with 28% undergoing SG and 10% undergoing AGB. Weight loss was higher for RYGB than SG (p<0.001) and AGB (p<0.00001). Too few records were available to allow statistical comparisons at five years.
Limitations
The study did not randomize patients to surgery type, and patient records did not have consistently collected BMI measurements. AGB data may be underrepresented because the data set did not include small ambulatory surgery centers.
Conclusions and Relevance
RYGB was associated with greater weight loss and diabetes remission than AGB or SG. AGB was associated with the lowest weight loss and least diabetes remission and the highest risk for additional intervention. SG had the lowest rate of additional intervention.
Future Research Needs
Future research could collect BMI data at set intervals.
| How this project fits under PCORI’s Research Priorities The PCORnet® Study reported in this results summary was conducted using PCORnet®, the National Patient-Centered Clinical Research Network. PCORnet® is intended to improve the nation’s capacity to conduct health research, particularly comparative effectiveness research (CER), efficiently by creating a large, highly representative network for conducting clinical outcomes research. PCORnet® has been developed with funding from the Patient-Centered Outcomes Research Institute® (PCORI®). |
More Results
Comparing Two Types of Weight Loss Surgery in Patients with or without Depression and by Race and Ethnicity -- A PCORnet® Study
⮟ View Professional Abstract ⮟ View Peer-Review Summary
Results Summary
What was the research about?
Some adults with severe obesity can’t lose enough weight with diet and exercise to improve their health. Surgery may help people lose weight and improve problems related to obesity like diabetes.
Mental health issues, like depression, may affect how well surgery works. Questions also remain about whether health outcomes after surgery differ by patients’ race and ethnicity.
In this study, the research team compared two types of weight loss surgery for people with or without depression and for people of different racial and ethnic groups: Roux-en-Y gastric bypass, or RYGB, and sleeve gastrectomy, or SG. The team looked at patients’ health records for up to five years to see how surgery affected:
- Weight loss in 36,871 patients with body mass index greater than or equal to 35
- Diabetes in 8,407 patients with type 2 diabetes
- Safety outcomes in 32,853 patients receiving care at health systems that could link health records with information about safety events and death
What were the results?
Comparing patients with and without depression. After either surgery, patients with or without depression didn’t differ in:
- Weight loss
- Improved blood sugar levels
- Whether diabetes came back after surgery
Patients with depression had a higher risk of having an endoscopy after surgery to detect problems than patients without depression. Rates of hospital stays were greatest for patients with depression who received RYGB. They were lowest for patients without depression who received SG.
Comparing Black, Hispanic, and White patients. Across racial and ethnic groups, compared with patients who had SG, patients who had RYGB:
- Lost more weight
- Had improved blood sugar levels and lower rates of diabetes coming back
- Had a higher risk of having an endoscopy or additional surgery
Black and Hispanic patients had a higher risk of having another hospital stay with RYGB than SG; White patients didn’t.
What did the research team do?
The research team reviewed health records from 25 health systems around the country for three groups of patients ages 20–79. All had RYGB or SG between January 2005 and September 2015.
- Of patients in the weight loss group, 53 percent were White, 24 percent were Hispanic, 19 percent were Black, and 2 percent were another race.
- Of patients in the diabetes group, 50 percent were White, 27 percent were Hispanic, 17 percent were Black, and 5 percent were another race.
- Of patients in the safety group, 52 percent were White, 26 percent were Hispanic, 18 percent were Black, and 2 percent were another race.
Health records didn’t have race or ethnicity for 2 percent of patients in all three groups.
Patients, surgeons, advocacy group members, and insurers helped design the study.
What were the limits of the study?
The study didn’t assign patients by chance to type of surgery. Other factors, like access to insurance or healthy food, may have affected the results.
Future research could explore how such factors affect health outcomes among patients who have weight loss surgery.
How can people use the results?
Patients with obesity and their doctors can use the results when considering weight loss surgery.
| How this project fits under PCORI’s Research Priorities The PCORnet® Study reported in this results summary was conducted using PCORnet®, the National Patient-Centered Clinical Research Network. PCORnet is intended to improve the nation’s capacity to conduct health research, particularly comparative clinical effectiveness research (CER), efficiently by creating a large, highly representative network for conducting clinical outcomes research. PCORnet has been developed with funding from PCORI®. |
Professional Abstract
Objective
To determine whether weight loss, type 2 diabetes, and safety outcomes from Roux-en-Y gastric bypass (RYGB) or sleeve gastrectomy (SG) differ by depression status or race and ethnicity
Study Design
| Design Elements | Description |
|---|---|
| Design | Observational: cohort study |
| Population | Patients ages 20–79 at the time of surgery in three cohorts: Diabetes outcomes cohort: 8,407 patients who had type 2 diabetes and a BMI of 35 or greater in the year before surgery Safety outcomes cohort: 32,853 patients from health systems who could link their electronic health records to (1) insurance claims or (2) state or national death indices that could track safety outcomes |
| Interventions/ Comparators |
|
| Outcomes | Percentage of total weight loss, change in HbA1c, diabetes relapse or remission, safety outcomes (reoperation, hospitalization, endoscopy, death) |
| Timeframe | 5-year follow-up for study outcomes |
This retrospective observational cohort study used data from 25 health systems to identify patients who had a preoperative body mass index (BMI) of 35 or greater and underwent primary bariatric surgery between 2005 and 2015. Using health records, researchers identified patients with or without depression prior to surgery. Researchers examined outcomes related to weight loss, diabetes, and safety for patients with or without depression up to five years after surgery for three cohorts:
- The weight loss cohort included 36,871 patients. Of these, 53% were White, 24% were Hispanic, 19% were Black, and 2% were another race or ethnicity.
- The diabetes outcomes cohort included 8,407 patients. Of these, 50% were White, 27% were Hispanic, 17% were Black, and 5% were another race or ethnicity.
- The safety outcomes cohort included 32,853 patients. Of these, 52% were White, 26% were Hispanic, 18% were Black, and 2% were another race or ethnicity.
Researchers used health records to identify patients’ race and ethnicity. Health records did not record this information for 2% of patients in all three cohorts.
Patients, surgeons, advocacy group members, and insurers helped design the study.
Results
Comparing patients with or without depression
- Weight loss cohort. Patients with or without depression did not differ significantly in weight loss after undergoing RYGB or SG.
- Diabetes outcomes cohort. Patients with depression had greater HbA1c improvements after RYGB than patients without depression (p<0.001); however, this difference was not clinically meaningful. Patients with or without depression who received SG did not differ significantly in HbA1c improvement. Following either surgery, the change in the diabetes remission or relapse did not differ significantly between patients with or without depression.
- Safety outcomes cohort. For either surgery, patients with depression had a higher risk of having an endoscopy to diagnose complications after surgery than patients without depression (p=0.003).
Comparing Black, Hispanic, and White Patients
Across racial and ethnic groups, compared with patients who had SG, patients who had RYGB:
- Lost more weight (p<0.001). This difference was greatest for White patients but was not clinically meaningful.
- Had greater reductions in HbA1c and lower rates of diabetes relapse (both p<0.05).
- Had a higher risk of having endoscopy or additional surgery (both p<0.001).
Black and Hispanic patients had a higher risk of hospitalization with RYGB than SG, but White patients did not.
Limitations
The study had a nonrandomized design. Confounders, like access to insurance and healthy food, may have affected the results.
Conclusions and Relevance
The study did not find clinically meaningful differences in weight loss, diabetes, or safety outcomes by depression status or race and ethnicity following bariatric surgery. Across cohorts, RYGB had greater improvements in weight loss and diabetes outcomes but higher risk for endoscopy than SG.
Future Research Needs
Future research could explore other characteristics, like access to insurance or healthy food, that may affect outcomes after surgery.
| How this project fits under PCORI’s Research Priorities The PCORnet® Study reported in this results summary was conducted using PCORnet®, the National Patient-Centered Clinical Research Network. PCORnet is intended to improve the nation’s capacity to conduct health research, particularly comparative clinical effectiveness research (CER), efficiently by creating a large, highly representative network for conducting clinical outcomes research. PCORnet has been developed with funding from PCORI®. |
Peer-Review Summary
Peer review of PCORI-funded research helps make sure the report presents complete, balanced, and useful information about the research. It also assesses how the project addressed PCORI’s Methodology Standards. During peer review, experts read a draft report of the research and provide comments about the report. These experts may include a scientist focused on the research topic, a specialist in research methods, a patient or caregiver, and a healthcare professional. These reviewers cannot have conflicts of interest with the study.
The peer reviewers point out where the draft report may need revision. For example, they may suggest ways to improve descriptions of the conduct of the study or to clarify the connection between results and conclusions. Sometimes, awardees revise their draft reports twice or more to address all of the reviewers’ comments.
Peer reviewers commented and the researchers made changes or provided responses. Those comments and responses included the following:
- Since this report describes an extension of the original study, the reviewers mostly agreed with the researchers, referring readers to the main final research report or prior publications for methods of the original interventions and analyses. There were some components of the statistical analysis plan that needed to be included. The researchers responded by adding information about the analytic assumptions made, how missing data were managed and how they dealt with the possibility of confounding.
- The reviewers asked the researchers to expand on their rationale for this research project extension and the specific hypotheses. The researchers expanded their presentation of their hypotheses to explain the rationale behind the hypotheses and the predicted differences.
Final Research Report
View this project's final research report.
Engagement Resources
Evidence Updates
Implementation
Related PCORI Dissemination and Implementation Project
Journal Citations
Article Highlights: The PCORnet® Bariatric Study, which is using data from PCORnet, The National Patient-Centered Clinical Research Network, to study outcomes among patients who undergo common weight-loss surgeries, has published several papers in prominent medical journals. The most recent clinical findings, published in JAMA Surgery, are on diabetes remission outcomes among patients who underwent gastric bypass and gastric sleeve surgeries. Another article, also published in JAMA Surgery, details the five-year outcomes of patients who underwent either gastric bypass or adjustable banding, finding that gastric bypass patients were significantly more likely than gastric sleeve patients to end up back in the hospital in the years following surgery.
Results of This Project
Related Journal Citations
Stories and Videos
Peer-Review Summary
Peer review of PCORI-funded research helps make sure the report presents complete, balanced, and useful information about the research. It also assesses how the project addressed PCORI’s Methodology Standards. During peer review, experts read a draft report of the research and provide comments about the report. These experts may include a scientist focused on the research topic, a specialist in research methods, a patient or caregiver, and a healthcare professional. These reviewers cannot have conflicts of interest with the study.
The peer reviewers point out where the draft report may need revision. For example, they may suggest ways to improve descriptions of the conduct of the study or to clarify the connection between results and conclusions. Sometimes, awardees revise their draft reports twice or more to address all of the reviewers’ comments.
Peer reviewers commented and the researchers made changes or provided responses. Those comments and responses included the following:
- The reviewers asked why the researchers did not perform propensity score analyses on some specific parts of the study, such as safety outcomes, to account for any differences between groups in this observational study. The researchers explained that they found the propensity score adjustments did not change results enough to warrant accepting certain assumptions and checking propensity distributions. In addition, because the hypotheses in these studies were associative instead of causal, the reviewers felt that the covariate adjustment was sufficient without a propensity score adjustment.
- The reviewers asked whether the researchers specified subgroup analyses in advance and whether they were driven by hypotheses. The researchers responded that they specified all subgroup analyses in advance but that they considered the analyses exploratory rather than driven by hypotheses.
- The reviewers asked why in the qualitative study, patient feedback was sought in focus groups rather than in individual interviews. Since obesity and bariatric surgery can be socially sensitive topics, the reviewers noted that having to answer questions in a group setting may have affected responses. The researchers said funding constraints did not allow for individual interviews, particularly since they wanted to maximize sample size and representation of patients from different parts of the country. They acknowledged that using focus groups was a limitation of the study.
- The reviewers asked why the frequency of obesity-related cancers, including colorectal cancer, was not analyzed as an outcome. The researchers explained that they excluded patients with certain types of cancers, especially gastrointestinal (GI) cancers, because some patients with GI cancer will undergo a procedure for the cancer that is coded the same as bariatric surgery in electronic health records.
Conflict of Interest Disclosures
Project Information
Key Dates
Study Registration Information
^Kaiser Permanente acquired Group Health Cooperative in February 2017.