The Breakfast Study

More than 15% of U.S. adults with type 2 diabetes have poorly controlled blood glucose, here defined as a glycated hemoglobin (HbA1c) level of 7.0% or higher. These adults have an elevated health risk of a variety of outcomes, including amputation and mortality from cardiovascular disease and from all causes. Nutrition- focused interventions can be effective for improving glycemic control, reducing anti-hyperglycemic medications, and reducing body weight, all of which are critical outcomes for adults with type 2 diabetes. However, typical nutrition-focused interventions can be burdensome, often requiring complex instructions and a complete overhaul of one's diet. Additionally, adults with poorly controlled type 2 diabetes are more likely to have low literacy levels, which can be a barrier for adherence to complex interventions. Therefore, an effective intervention for adults with poorly controlled diabetes who may have lower health literacy levels is necessary to reduce both HbA1c levels and anti-hyperglycemic medications. Carbohydrate intake has the strongest impact on post-prandial glycemia of any dietary factor, and a very low-carbohydrate diet-due to its ability to improve glycemic control-is now recommended by the American Diabetes Association (ADA) for the treatment of type 2 diabetes. The investigators hypothesized that some of the benefits of a very low-carbohydrate diet may be available to individuals who change only their breakfasts to be very low-carbohydrate, rather than modifying their entire diet.

Thus, the investigators conducted an acceptability, feasibility, preliminary effectiveness trial of a 4-month, online, small- steps, low-literacy, very low-carbohydrate breakfast-focused program in 120 adults with poorly controlled type 2 diabetes. The investigators measured acceptability and feasibility, plus critical efficacy outcomes, such as changes in HbA1c, anti-hyperglycemic medications, glycemic variability, body weight, blood pressure, and lipids.

The investigators also tested whether factors such as sex, health literacy level, and baseline insulin resistance significantly moderate the impact of the intervention on change in HbA1c and change in anti-hyperglycemic medications.

• Breakfast (Behavioral): Participants were provided recipes and information to support this dietary change.

Primary outcomes

• Intervention Satisfaction [At 4 months]: A 1-item measure about intervention satisfaction. We asked participants, "How would you rate your overall satisfaction with the program?" The item was rated from 1 = very dissatisfied to 6 = very satisfied, with higher scores reflecting greater satisfaction.
• Assessment of Program Feasibility [At Month 4]: For feasibility, the percentage of completed sessions out of 16 possible sessions was calculated for each participant, and the mean percentage across participants was reported.

Secondary outcomes

• Weekly Dietary Adherence [Weekly over 4 months]: A 1-item measure about dietary adherence. We asked participants to reflect on the past week and rate whether they ate a low-carbohydrate breakfast. The item was rated from 1 = not at all to 7 = very much so, with higher scores reflecting greater adherence. Dietary adherence was collected weekly over the 4-month intervention. Average adherence across the intervention was reported.
• Change in PROMIS Sleep Disturbance T-score [Baseline, 4 months]: The 4-item PROMIS sleep disturbance subscale measures the extent to which patients experience problems with sleep over the past 7 days using a 5-point Likert scale. The subscale is measured using T-scores (mean 50, SD 10), with higher scores reflecting more severe sleep problems. To assess the effect of dietary change on sleep disturbance, the change in PROMIS sleep disturbance T-scores from baseline to 4 months was evaluated.
• Change in PROMIS Pain Interference T-score [Baseline, 4 months]: The 4-item PROMIS subscale for pain interference measures the extent to which patients experience problems with pain interference over the past 7 days using a 5-point Likert scale. The subscale is measured using T-scores (mean 50, SD 10), with higher scores reflecting greater pain interference. To assess the effects of dietary changes on pain, the change in PROMIS pain interference T-scores from baseline to 4 months was evaluated.
• Change in PROMIS Anxiety T-score [Baseline, 4 months]: The 4-item PROMIS subscale for anxiety measures the extent to which patients experience depressive symptoms over the past 7 days using a 5-point Likert scale. The subscale is measured using T-scores (mean 50, SD 10), with higher scores reflecting greater depression. To assess the effects of dietary changes on anxiety, the change in PROMIS anxiety T-scores from baseline to 4 months was evaluated.
• Change in PROMIS Depression T-score [Baseline, 4 months]: The 4-item PROMIS subscale for depression measures the extent to which patients experience depressive symptoms over the past 7 days using a 5-point Likert scale. The subscale is measured using T-scores (mean 50, SD 10), with higher scores reflecting greater depression. To assess the effects of dietary changes on depression, the change in PROMIS depression T-scores from baseline to 4 months was evaluated.
• Change in PROMIS Fatigue T-score [Baseline, 4 months]: The 4-item PROMIS subscale for fatigue measures the extent to which patients experience problems with fatigue over the past 7 days using a 5-point Likert scale. The subscale is measured using T-scores (mean 50, SD 10), with higher scores reflecting greater fatigue. To assess the effects of dietary changes on fatigue, the change in PROMIS fatigue T-scores from baseline to 4 months was evaluated.
• Change in PROMIS Physical Function T-score [Baseline, 4 months]: The 4-item PROMIS subscale for physical function measures the extent to which patients experience problems with physical function over the past 7 days using a 5-point Likert scale. The subscale is measured using T-scores (mean 50, SD 10), with higher scores reflecting better physical function. To assess the effects of dietary changes on physical function, the change in PROMIS physical function T-scores from baseline to 4 months was evaluated.
• Change in PROMIS Social Roles T-score [Baseline, 4 months]: The 4-item PROMIS subscale for social roles measures the extent to which patients experience problems with their social roles over the past 7 days using a 5-point Likert scale. The subscale is measured using T-scores (mean 50, SD 10), with higher scores reflecting better social roles. To assess the effects of dietary changes on social roles, the change in PROMIS social roles T-scores from baseline to 4 months was evaluated.
• Change in PROMIS Pain Intensity [Baseline, 4 months]: The 1-item PROMIS subscale for pain intensity measures average pain intensity over the past 7 days using a 10-point Likert scale, with higher scores reflecting greater pain. To assess the effects of dietary changes on pain intensity, the change in PROMIS pain intensity score from baseline to 4 months was evaluated.
• Change in HbA1c [Baseline, 4 months]: HbA1c is the proportion of hemoglobin in the blood that has glucose attached to it relative to total hemoglobin. It reflects average blood glucose levels over approximately the previous 2-3 months. To assess the effect of dietary change on HbA1c, the change in HbA1c from baseline to 4 months were evaluated. A negative change indicates an improvement in HbA1c from baseline.
• Change in Anti-hyperglycemic Medications [Baseline, 4 months]: The Medication Effect Score (MES) quantifies the overall intensity of antiglycemic therapy based on potency and dosages of medications (Alexopoulos et al., 2021), with higher scores indicating greater medication intensity. It is interpreted as the reduction in HbA1c attributable to the medication regime. For each diabetes medication, the percentage of the maximum recommended daily dose is multiplied by an adjustment factor reflecting its expected hemoglobin A1c-lowering potency; these values are summed to yield the total MES. Scores range from 0 upward (no fixed maximum), with lower scores reflecting lower medication intensity. The change in MES from baseline to 4 months was evaluated, with negative values indicating a reduction in medication intensity.

Sex: All, ages: 18 Years to 80 Years

Inclusion Criteria:

* HbA1c 7% or higher
* 18-80 years old
* The ability to understand verbal and written English
* Willingness to follow the prescribed diet
* Able to consent and follow directions
* Willingness to regularly check blood glucose levels as required

Exclusion Criteria:

* Inability to provide informed consent
* Pregnant, breastfeeding, or planning for either in the next 6 months or <6 months postpartum
* Low C-peptide and possible subsequent GAD 65 level that suggests type 1 diabetes, clinical factors that suggest type 1 diabetes (lean, lack of family history, and diabetic ketoacidosis in the past) or a previous diagnosis of type 1 diabetes or latent autoimmune diabetes
* Triglycerides are greater than 999mg/dL on the screening blood draw
* Cancer, heart failure, or kidney failure
* Vegan
* Untreated mental health condition
* Currently following a very low-carbohydrate diet or breakfasts
* Thyroid levels out of range
* Alcoholism
* Previous bariatric surgery
* Difficulty chewing or swallowing
* Dependence on others for food preparation
* Currently enrolled in another investigative study that might conflict with this research

Participant flow

Outcome results

• Intervention Satisfaction (Primary), Mean, score on a scale: Very Low-carbohydrate Breakfast 5.44 +/- 0.91 (p <0.001)
• Assessment of Program Feasibility (Primary), Mean, percentage of sessions completed: Very Low-carbohydrate Breakfast 90.20 +/- 17.59 (p <0.001)
• Weekly Dietary Adherence (Secondary), Mean, score on a scale: Very Low-carbohydrate Breakfast 6.3 +/- 1.1
• Change in PROMIS Sleep Disturbance T-score (Secondary), Mean, T-score: Very Low-carbohydrate Breakfast -0.2 (-1.6 to 1.2) (p 0.761)
• Change in PROMIS Pain Interference T-score (Secondary), Mean, T-score: Very Low-carbohydrate Breakfast -0.0 (-1.4 to 1.3) (p 0.967)
• Change in PROMIS Anxiety T-score (Secondary), Mean, T-score: Very Low-carbohydrate Breakfast -0.7 (-2.0 to 0.6) (p 0.290)
• Change in PROMIS Depression T-score (Secondary), Mean, T-score: Very Low-carbohydrate Breakfast -0.9 (-2.0 to 0.2) (p 0.115)
• Change in PROMIS Fatigue T-score (Secondary), Mean, T-score: Very Low-carbohydrate Breakfast -0.8 (-2.2 to 0.6) (p 0.240)
• Change in PROMIS Physical Function T-score (Secondary), Mean, T-score: Very Low-carbohydrate Breakfast -0.4 (-1.3 to 0.5) (p 0.344)
• Change in PROMIS Social Roles T-score (Secondary), Mean, T-score: Very Low-carbohydrate Breakfast 0.8 (-0.4 to 1.9) (p 0.180)
• Change in PROMIS Pain Intensity (Secondary), Mean, units on a scale: Very Low-carbohydrate Breakfast 0.4 (0.0 to 0.8) (p 0.029)
• Change in HbA1c (Secondary), Mean, % of hemoglobin with glucose (HbA1c): Very Low-carbohydrate Breakfast -0.4 (-0.6 to -0.2) (p <0.001)
• Change in Anti-hyperglycemic Medications (Secondary), Mean, MES units: Very Low-carbohydrate Breakfast -0.2 (-0.4 to -0.1) (p <0.001)

Adverse events

Time frame: 16 weeks

Serious adverse events by term

Other adverse events by term

• Results publication: Hansen A, Raymond K, Simon A, Kim S, Bayandorian H, Marriott D, Saslow LR. Very Low-Carbohydrate Breakfast Intervention for Adults with Type 2 Diabetes and Persistent Hyperglycemia: Protocol for a Digital, Nonrandomized Pre-Post Study. JMIR Res Protoc. 2026 Jan 28;15:e81041. doi: 10.2196/81041. PMID 41605502

• Study Protocol (PDF, 2026-10-01)
• Statistical Analysis Plan (PDF, 2026-09-11)
• Informed Consent Form (PDF, 2024-11-14)

NIH grants: R01DK132099