Results Summary
What was the research about?
Losing weight may prevent or delay the start of type 2 diabetes and prevent health problems due to diabetes. Obesity counseling called intensive behavioral therapy, or IBT, is effective for weight loss. In IBT, doctors screen and counsel patients about weight management. After 2012, health insurers started to reimburse primary care doctors for IBT.
In this study, the research team used health records and insurance claims from 2009 to 2020 for patients with diabetes or at risk for diabetes at six health systems to see how:
- The payment change for doctors affected use of IBT.
- IBT affected patients’ weight loss and other health outcomes.
What were the results?
Before the payment change, health records rarely captured IBT use. After the payment change, less than 1 percent of eligible patients received IBT. Among patients receiving IBT, the usual number of IBT visits was one.
Among patients with diabetes, weight loss and blood pressure control didn’t differ between patients who did and didn’t receive IBT. Patients who received IBT had a larger decrease in blood sugar levels. They also had a smaller increase in uncontrolled diabetes.
Among patients at risk for diabetes, patients gained weight whether they received IBT or not. Patients who received IBT gained more weight than patients who didn't receive it. Changes in blood sugar levels and blood pressure didn’t differ between patients who did and didn’t receive IBT.
What did the research team do?
The research team looked at how many patients received IBT before and after the payment change. The team looked at data for 567,908 patients with diabetes and 2,054,256 patients at risk for diabetes. Across these patients, 80 percent were White, 11 percent were Black or African American, 8 percent were another race or race was missing; 5 percent were Hispanic. The average age was 47, and 57 percent were women.
To assess the effects of IBT, the research team compared patients who received IBT with patients who didn’t receive it but were eligible based on their weight. The team looked at changes in health outcomes from one year before patients had, or were eligible for, IBT versus one year later for:
- 4,944 patients with diabetes. Of these, 75 percent were White, 20 percent were Black or African American, and 1 percent were another race; 3 percent were Hispanic. The average age was 58, and 62 percent were women.
- 10,781 patients at risk for diabetes. Of these, 83 percent were White, 12 percent were Black or African American, and 1 percent were another race; 3 percent were Hispanic. The average age was 44, and 80 percent were women.
Patients with diabetes, doctors, and staff from health systems, diabetes organizations, and community organizations gave input on the study.
What were the limits of the study?
Data didn’t include things like doctor practices, which could have affected the results.
Future research could look at reasons for the low use of IBT.
How can people use the results?
Patients and their doctors can use the results when considering IBT for diabetes.
| How this project fits under PCORI’s Research Priorities The research 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
(1) To evaluate the effect of primary care provider reimbursement on intensive behavioral therapy (IBT) utilization; (2) To assess the effect of IBT on weight loss and other health outcomes among patients with or at risk for type 2 diabetes
Study Design
| Design Element | Description |
|---|---|
| Design |
|
| Population |
Electronic health record and insurance claims data from the PaTH Clinical Data Research Network, which included 6 health systems in Pennsylvania, Maryland, and Utah Quasi-experimental study included 2009–2020 data for:
Cohort study included data for:
|
| Interventions/ Comparators |
|
| Outcomes |
Quasi-experimental study: IBT utilization Cohort study: weight loss (primary); HbA1c, controlled blood pressure, uncontrolled diabetes (secondary) |
| Timeframe |
Quasi-experimental study: Up to 8-year follow-up from policy change Cohort study: 1-year follow-up for study outcomes |
Researchers conducted two studies examining IBT utilization and its effect on health outcomes among patients with or at risk for diabetes. In IBT, doctors counsel patients about weight management. After 2012, the Centers for Medicare & Medicaid Services (CMS) and private insurers reimbursed primary care providers for IBT.
The quasi-experimental study examined IBT utilization by determining the percentage of eligible patients whose providers billed for IBT services in two cohorts:
- 567,908 patients with diabetes
- 2,054,256 patients at risk for diabetes
Across both cohorts, 80% were White, 11% were Black or African American, 8% reported other race or race was missing; 5% were Hispanic. The average age was 47, and 57% were female.
In the cohort study, researchers matched each patient who received IBT to a randomly selected patient who was eligible but did not receive IBT. Researchers compared changes in outcomes from one year prior to one year after the first IBT visit or the first IBT-eligible visit. The two cohorts included:
- 4,944 patients with diabetes. Of these, 75% were White, 20% were Black or African American, 1% were another race; 2% were Hispanic. The average age was 58, and 62% were women.
- 10,781 patients at risk for diabetes. Of these, 83% were White, 12% were Black or African American, 1% were another race; 3% were Hispanic. The average age was 44, and 80% were women.
Patients with diabetes, doctors, and representatives from health systems provided input during the study.
Results
Quasi-experimental study. Prior to the start of reimbursement, electronic health records rarely recorded IBT services. After reimbursement, few eligible patients received IBT: 0.6% of the type 2 diabetes cohort and 0.3% of the at-risk cohort. Among patients who received IBT, the median number of IBT visits was one.
Cohort study. Among patients with diabetes, patients who did and did not receive IBT did not differ significantly in weight loss or blood pressure control. Patients who received IBT had a larger decrease in hemoglobin A1C (HbA1c) levels (-0.21; p<0.001) and a smaller increase in uncontrolled diabetes (0.19; p=0.01) than patients who did not receive it.
Among patients at risk for diabetes, patients gained weight whether they received IBT or not; patients receiving IBT experienced more weight gain (5.24 pounds versus 1.75 pounds; p<0.001). Changes in HbA1C and blood pressure control did not differ significantly among patients who did and did not receive IBT.
Limitations
Data did not include variables like clinician practices, which could have affected results.
Conclusions and Relevance
IBT utilization was low among providers caring for patients with or at risk for diabetes; weight loss did not differ significantly between patients who received and did not receive IBT.
Future Research Needs
Future research could explore reasons for low IBT utilization in primary care.
| How this project fits under PCORI’s Research Priorities The research 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®). |
COVID-19-Related Study
Assessing the Effects of Telehealth Use and Identifying Risks for Severe COVID-19 among Patients with or at Risk for Type 2 Diabetes
Results Summary
In response to the COVID-19 public health crisis in 2020, PCORI launched an initiative to enhance existing research projects so that they could offer findings related to COVID-19. The initiative funded this study and others.
What was this COVID-19 study about?
People with type 2 diabetes are at high risk for severe illness or death from COVID-19. During the COVID-19 pandemic, people with or at risk for diabetes often had to shift from in-person care to telehealth. Telehealth provides care to patients remotely using phone or video. Looking at the effects of telehealth and risks for developing severe COVID-19 may help inform care for these patients.
The research team did two studies using health records and insurance claims for patients with or at risk for type 2 diabetes:
- Study 1 looked at the effects of telehealth on patients' blood sugar levels and other health outcomes.
- Study 2 looked at risks for a hospital stay for COVID-19.
What were the results?
Study 1. Among patients with diabetes, changes in blood sugar levels didn’t differ between patients who did and didn’t use telehealth during the pandemic.
The research team also looked at patients with or at risk for diabetes who tested positive for COVID-19. Among these patients, patients who used telehealth were less likely to need a hospital stay than patients who didn’t use telehealth. Patients who used telehealth were also less likely to need a ventilator or intensive care, or to die within 30 days of a COVID-19 diagnosis.
Study 2. Among patients with COVID-19 and diabetes, patients were more likely to need a hospital stay if they:
- Were Hispanic compared with those who were White.
- Were over age 65.
- Had other long-term health problems.
- Took insulin.
Among patients with COVID-19 who were at risk for diabetes, patients were more likely to need a hospital stay if they:
- Were Hispanic compared with those who were White.
- Had other long-term health problems.
What did the research team do?
Study 1. The research team reviewed health records and insurance claims from March 2019 to February 2021. The study used data for 1,284,457 patients with or at risk for type 2 diabetes and receiving care in two states. Among patients, 80 percent were White, 12 percent were Black, and 4 percent were another race or race was missing; 4 percent were Hispanic. The average age was 56, and 57 percent were women.
Study 2. The study used data from 15,725 patients from study 1 who had COVID-19. Of these, 60 percent were White, 23 percent were Black, and 5 percent were another or unknown race; 11 percent were Hispanic. The average age was 57, and 60 percent were women. The research team used statistical methods to look at the effect of risk factors on getting severe COVID-19.
Patients with diabetes, doctors, policy makers, and staff from health systems, diabetes organizations, and community agencies gave input on both studies.
What were the limits of the study?
Health records often have missing data. Missing data for blood sugar levels could have affected the results.
How can people use the results?
Doctors and health systems can use the results when considering ways to care for patients with or at risk for diabetes.
| The research 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
In response to the COVID-19 public health crisis in 2020, PCORI launched an initiative to enhance existing research projects so that they could offer findings related to COVID-19. The initiative funded this study and others.
Background
Type 2 diabetes is a risk factor for increased COVID-19 mortality. During the pandemic, patients with or at risk for diabetes relied on telehealth visits for outpatient care. Examining the effects of telehealth and identifying factors associated with developing severe COVID-19 may help inform care for these patients.
Objective
(1) To understand how telehealth visits affected hemoglobin A1c (HbA1c) levels for patients with type 2 diabetes; (2) To identify factors associated with severe COVID-19 among patients with and at risk for type 2 diabetes
Study Design
| Design Element | Description |
|---|---|
| Design |
|
| Population |
Cohort study: 1,284,457 patients including 286,961 adult patients with type 2 diabetes and a body mass index (BMI) ≥30 kg/m2, and 997,496 patients at risk for type 2 diabetes based on BMI ≥25 kg/m2 with at least 1 ambulatory care visit between March 1, 2019, and February 29, 2020 Descriptive study: 15,725 adult patients with type 2 diabetes and BMI ≥30 kg/m2 or at risk for type 2 diabetes based on BMI ≥25 kg/m2 with at least 1 ambulatory care visit between March 1, 2019, and February 29, 2020, and diagnosed with COVID-19 |
| Outcomes |
Cohort study: HbA1c levels (primary), healthcare utilization (secondary) Descriptive study: COVID-19 severity (using odds of hospitalization as a proxy measure) |
| Data Collection Timeframe |
Cohort study: pre-pandemic period of March 2019–February 2020; pandemic period of March 2020–February 2021 Descriptive study: March 2020–February 2021 |
The retrospective cohort study compared the effect of using telehealth before versus during the pandemic on changes in HbA1c levels among patients with diabetes. The study included health records and insurance claims data from 1,284,457 patients with or at risk for diabetes in two states. Among patients, 80% were White, 12% were Black, and 4% were another race or race was missing; 4% were Hispanic. The average age was 56, and 57% were female.
The descriptive study assessed the association between patient characteristics and severe COVID-19. Researchers used data from the cohort study for 15,725 patients with COVID-19 with or at risk for diabetes. Of these, 60% were White, 23% were Black, 5% were another or unknown race; 11% were Hispanic. The average age was 57, and 60% were female.
Patients with diabetes, doctors, and representatives from health systems, professional organizations, and community organizations provided input during the study.
Results
Cohort study. Among patients with diabetes, changes in HbA1c levels did not differ significantly between patients who did and did not receive telehealth.
Among patients with or at risk for diabetes who tested positive for COVID-19, patients who received telehealth had lower odds of:
- Hospitalization than patients without telehealth (patients with diabetes odds ratio [OR]=0.73; 95% confidence interval [CI]: 0.61, 0.89; patients at risk for diabetes OR=0.60; 95% CI: 0.52, 0.69).
- Needing a ventilator or intensive care, or of dying within 30 days of COVID-19 diagnosis (patients with diabetes OR=0.70; 95% CI: 0.50, 0.90; patients at risk for diabetes OR=0.63; 95% CI 0.5, 0.8).
Descriptive study. Among patients with COVID-19 and diabetes, odds of hospitalization increased for patients who were Hispanic compared to White (OR=1.52; 95% CI: 1.19, 1.93), were over age 65 (OR=1.26; 95% CI 1.04, 1.53), had more comorbid conditions (OR=2.09; 95% CI: 1.79, 2.44), or took insulin (OR=3.31; 95% CI: 2.77, 3.94). HbA1c level was not significantly associated with hospitalization.
Among patients with COVID-19 at risk for diabetes, odds of hospitalization increased for patients who were Hispanic compared to White (OR=1.78; 95% CI 1.44, 2.19) and who had more comorbid conditions (OR=4.55; 95% CI: 3.91, 5.28).
Limitations
Missing data for HbA1c in health records could have affected results.
Conclusions and Relevance
In this study, telehealth did not affect HbA1c levels among people with diabetes and may have a protective effect among patients with COVID-19 and diabetes or at risk for diabetes. Certain populations, such as people with comorbid conditions, had higher odds of hospitalization for COVID-19.
| The research 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®). |
Peer Review Summary
The Peer-Review Summary for this COVID-19-related study will be posted here soon.
Final Enhancement Report
This COVID-19 study's final enhancement report is expected to be available by February 2023.
Final Research Report
View this project's final research report.
Engagement Resources
Journal Citations
Results of This Project
Related Journal Citations
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:
- This final report was a combined description of the main research study and the COVID-19 enhancement project. Peer reviewers were asked to assess both parts of the work.
- The statistical reviewer advised the researchers to provide more information on how they calculated sample size for the study, and in particular to clarify what the researchers meant in stating that their sample size could detect “very small effect sizes.” The researchers expanded their sample size justification by describing how they estimated the sample sizes in the planning stage of the study. In addition the researchers specified that they calculated that their sample size would have 90% power to detect an effect as small as 0.026 units of change in the main outcomes.
- Reviewers noted that the COVID-19 enhancement project (aim 3 in the report) was unclear on whether the aim was focused on telemedicine access or telemedicine use. The researchers edited the report to clarify that the study focused telemedicine use and identified additional limitations to making conclusions about the relationship between telemedicine use and the severity of COVID-19 infection. The reviewers cautioned the researchers to also avoid any language in the report that might lead readers to think that telemedicine use had a causal link to COVID-19 severity as opposed to having an association based on the study results.
- The reviewers suggested that the researchers provide more justification for comparing a single intensive behavioral therapy visit to no visit, asking whether going to one visit would be likely to have a clinically significant effect on weight loss. The researchers responded that since there is no evidence of how many sessions are associated with clinically significant weight loss, they analyzed the study participants based on how many treatment visits participants made. Since more than half of participants had only one session and there was no evidence that more sessions were associated with greater weight loss, the researchers decided to use one visit as the threshold.