An Almond-Based Low Carbohydrate Diet Improves Depression and Glycometabolism in Patients with Type 2 Diabetes through Modulating Gut Microbiota and GLP-1: A Randomized Controlled Trial

The researchers compared two diets head-to-head:

**Intervention:** An almond-based low-carbohydrate diet (a-LCD). This was a diet where carbohydrates were restricted (exact carb target not specified in the abstract, but typical LCD protocols aim for 20–50g net carbs per day), and almonds were a key component. Participants received specific almond portions as part of their daily meal plan.

**Comparator:** A low-fat diet (LFD). This is the standard dietary recommendation for type 2 diabetes management, focusing on reducing total fat intake, particularly saturated fat.

**Primary outcomes:**

1. Depression severity, measured by a validated questionnaire.

2. Glycometabolism, measured primarily by glycosylated hemoglobin (HbA1c) — a 3-month average of blood sugar levels.

**Secondary outcomes:**

1. Gut microbiota composition (types and abundance of bacteria in stool samples).

2. Fasting glucagon-like peptide 1 (GLP-1) concentration — a hormone that stimulates insulin release, slows digestion, and promotes satiety.

**Sample size:** 45 participants completed the study (22 in the almond-LCD group, 23 in the low-fat diet group). The study started with more participants, but some dropped out — the exact number randomized is not given in the abstract.

**Population:** Adults diagnosed with type 2 diabetes mellitus (T2DM).

**Setting:** Recruited from a diabetes club and the Endocrine Division of the First and Second Affiliated Hospital of Soochow University, China.

**Duration:** December 2018 to December 2019.

**Specifics:** Chinese adults with T2DM. No age range, BMI range, or medication status is provided in the abstract, but these would be important details for assessing generalizability.

**Depression:** Measured using a depression scale. The abstract does not name the specific instrument, but common validated scales include the Beck Depression Inventory (BDI-II, 0–63 scale, higher = more depressed) or the Hamilton Depression Rating Scale (HAM-D). The paper likely used one of these.

**Glycometabolism:** Glycosylated hemoglobin (HbA1c) — measured as a percentage. Normal is <5.7%, prediabetes 5.7–6.4%, diabetes ≥6.5%. This reflects average blood glucose over ~3 months.

**Gut microbiota:** Stool samples were collected and analyzed using 16S rRNA gene sequencing. This identifies which bacterial species are present and their relative abundance.

**GLP-1:** Fasting blood samples were taken to measure fasting GLP-1 concentration (typically in pmol/L).

**Study design:** This was a randomized controlled trial (RCT) — the gold standard for testing causal effects of an intervention.

**Randomisation:** Participants were randomly assigned to either the almond-based low-carbohydrate diet or the low-fat diet. This is critical because it balances known and unknown confounding factors (age, sex, baseline depression, baseline blood sugar) between groups, so any difference at the end can be attributed to the diet rather than pre-existing differences.

**Blinding:** The abstract does not mention blinding. In dietary studies, true double-blinding is nearly impossible — participants know what they're eating. However, researchers could have been blinded to group assignment during data analysis. The lack of blinding is a major limitation because participants' expectations about the "special" almond diet could influence self-reported depression scores (placebo effect).

**Duration:** 3 months. This is a reasonable duration for seeing changes in HbA1c (which reflects the previous 3 months) and for gut microbiota to shift. However, depression changes can occur faster or slower depending on the individual.

**Dietary adherence:** The abstract does not describe how adherence was monitored (e.g., food diaries, regular check-ins, biomarker verification). This is a critical gap — if participants didn't actually follow the diets, the results are meaningless.

**Statistical approach:** The abstract reports p-values (e.g., p < 0.05) but does not give specific effect sizes, confidence intervals, or describe which statistical tests were used (t-tests, ANOVA, ANCOVA adjusting for baseline differences). This is a weakness — "p < 0.05" alone tells you the result is unlikely due to chance, but not how big the effect is or how precise the estimate is.

**What this design can and cannot prove:**

**Can prove:** That the almond-based low-carbohydrate diet caused a different change in depression scores and HbA1c compared to the low-fat diet over 3 months in this specific population.

**Cannot prove:** That almonds specifically (rather than carbohydrate restriction in general) caused the effect — because the intervention combined both almonds AND carb restriction. It also cannot prove the mechanism (gut microbiota → GLP-1 → depression) — that's a hypothesis based on correlations, not a proven causal pathway. The study cannot tell us about long-term effects beyond 3 months, effects in non-diabetic populations, or effects in people with clinical depression (vs. elevated depressive symptoms in diabetes).

**Major methodological weaknesses (from abstract alone):**

No blinding

No adherence data reported

Small sample size (45 completers)

Specific depression scale not named

No effect sizes or confidence intervals reported

Possible industry funding (almond growers? — not stated but worth checking)

Single geographic location (China) limits generalizability

The abstract reports that the almond-based low-carbohydrate diet group showed statistically significant improvements compared to the low-fat diet group, but provides almost no specific numbers. Based on the abstract text:

**Depression:** The a-LCD group had a significantly greater reduction in depression scores compared to the LFD group (p < 0.05). The actual mean change and between-group difference are not reported in the abstract.

**HbA1c:** The a-LCD group had a significantly greater reduction in HbA1c compared to the LFD group (p < 0.05). Again, no specific values given.

**Gut microbiota:** The a-LCD group showed increased abundance of short-chain fatty acid (SCFA)-producing bacteria (specific genera not named in abstract). This is a secondary outcome and should be interpreted as exploratory.

**GLP-1:** Fasting GLP-1 concentrations increased significantly more in the a-LCD group compared to the LFD group (p < 0.05). No specific values.

**Critical note:** The abstract is unusually sparse in reporting actual numbers. A full-text review would be needed to extract the actual effect sizes. For a self-experimenter, this lack of specificity is frustrating — we don't know if the depression improvement was 2 points on a 63-point scale (clinically trivial) or 10 points (clinically meaningful).

Because the abstract does not report actual numbers, we cannot translate the effect into plain English with confidence. However, based on typical findings in similar studies:

**HbA1c:** A low-carbohydrate diet in type 2 diabetes typically reduces HbA1c by 0.5–1.0 percentage points over 3 months. A reduction of 0.5% is considered clinically meaningful. The almond-based version might add an additional 0.2–0.3% benefit.

**Depression:** In people with diabetes, dietary interventions typically reduce depression scores by 20–30% on validated scales. This could mean a drop from moderate depression (score ~20 on BDI) to mild depression (score ~14), or from mild to minimal.

**Without the actual numbers, treat these as rough estimates only.** The paper's full text is needed for precise effect sizes.

**Acknowledged by authors (likely):**

Small sample size (n=45 completers)

Short duration (3 months)

Inability to determine whether almonds or carbohydrate restriction drove the effect

Lack of measurement of SCFAs and GPR43 activation (they mention this as a future direction)

Possible confounding by other lifestyle factors not controlled for

**Additional critical limitations:**

**No blinding:** Participants knew which diet they were on. The almond-based diet may have seemed more "special" or "experimental," leading to greater expectation of benefit (placebo effect), especially for self-reported depression.

**No adherence verification:** Without objective biomarkers (e.g., urinary ketones for carb restriction, or almond-specific biomarkers), we don't know if participants actually followed the diets.

**No effect sizes reported:** The abstract only gives p-values, which is poor scientific reporting. A p-value tells you the result is unlikely due to chance, but not whether the effect is large enough to matter.

**Single depression measurement:** Depression was measured at baseline and 3 months only. More frequent measurements would capture trajectory and variability.

**No control for medication changes:** People with diabetes often adjust medications (insulin, metformin) during dietary changes. If the almond-LCD group reduced medications while the LFD group did not, this could confound results.

**Population specificity:** Chinese adults with type 2 diabetes. Gut microbiota composition varies by ethnicity, diet, and geography. Results may not generalize to other populations.

**Industry funding not disclosed in abstract:** Almond-based studies are sometimes funded by almond industry boards. This does not invalidate the results but should be considered.

For someone running their own n=1 experiment:

### What to test

**Intervention:** A low-carbohydrate diet (aim for 30–50g net carbs per day) with 30–50g of almonds daily (about 20–30 almonds, or 2 small handfuls). Almonds can be eaten as a snack, added to salads, or ground into almond flour for baking.

**Alternative test:** If you want to isolate the almond effect, you could test a standard low-carb diet for 3 months, then add almonds for another 3 months. Or test almonds alone without carb restriction (though the paper combined both).

### Minimum meaningful duration

**3 months minimum.** HbA1c reflects the previous 3 months, so shorter durations won't capture the full blood sugar effect. Gut microbiota shifts can occur in 2–4 weeks, but stabilization takes longer. Depression changes may be noticeable earlier (2–4 weeks), but 3 months gives a more reliable picture.

**If you're impatient:** Measure fasting blood glucose and depression weekly. You might see trends within 2–4 weeks, but don't draw conclusions until 3 months.

### What to measure

**Primary metric 1 – Depression:** Use a validated scale weekly. The **Patient Health Questionnaire-9 (PHQ-9)** is free, validated, and widely used (0–27 scale, 0-4 none, 5-9 mild, 10-14 moderate, 15-19 moderately severe, 20-27 severe). Or use the **Beck Depression Inventory (BDI-II)** if you have access.

**Primary metric 2 – Blood sugar:** Measure **fasting blood glucose** daily (first thing in the morning, before eating). Also measure **HbA1c** at baseline and at 3 months (available via home test kits or lab).

**Secondary metrics:**

- **Fasting insulin** (optional, at baseline and 3 months) — to calculate HOMA-IR (insulin resistance).

- **Weight and waist circumference** weekly.

- **Energy levels and mood** daily (simple 1–10 scale).

- **Gut symptoms** (bloating, gas, bowel movement frequency) — almonds can cause GI issues in some people.

**Optional advanced:** Stool testing for gut microbiota (companies like Viome, Thryve, or uBiome offer at-home kits). This is expensive and the clinical utility for an individual is debated, but it would directly test the proposed mechanism.

### Key confounds to control for

**Medication changes:** If you take diabetes medications (metformin, insulin, sulfonylureas), low-carb diets often require dose reductions to prevent hypoglycemia. Work with your doctor. Document any medication changes — they could be the real cause of blood sugar improvements.

**Calorie intake:** Low-carb diets often reduce appetite, leading to unintentional calorie restriction. Weight loss itself improves blood sugar and mood. To isolate the almond/carb effect, try to keep calories constant (or at least track them).

**Exercise:** Keep your exercise routine constant throughout the experiment. A new workout program could independently improve mood and blood sugar.

**Sleep:** Poor sleep worsens blood sugar and mood. Track sleep quality (duration, consistency) and note any changes.

**Stress:** Major life events (job change, relationship stress, financial worries) can dramatically affect both depression and blood sugar. Keep a simple stress log (1–10 daily).

**Other dietary changes:** If you start eating more vegetables, fiber, or fermented foods alongside the almonds, those could be the active ingredients. Try to change only the almond and carb variables.

**Seasonal effects:** Depression can vary with seasons (SAD). If you start in winter and end in spring, improvement may be due to sunlight, not almonds.

### What a positive result would look like

**Depression:** A drop of **5+ points on PHQ-9** (e.g., from 15 to 10) or a drop of **10+ points on BDI-II** (e.g., from 25 to 15). This is considered a clinically meaningful improvement. A drop of 2–3 points is statistically detectable but may not feel life-changing.

**Blood sugar:** A drop in **HbA1c of 0.5 percentage points or more** (e.g., from 7.5% to 7.0% or lower). Fasting blood glucose should trend downward by 10–30 mg/dL (0.6–1.7 mmol/L).

**GLP-1 (if measured):** An increase of 20–50% in fasting GLP-1 levels.

**Gut microbiota (if measured):** Increased abundance of SCFA-producing genera like *Faecalibacterium*, *Roseburia*, *Lachnospira*, or *Bifidobacterium*.

**Timeline:** Blood sugar improvements may appear within 1–2 weeks (due to carb restriction) but stabilize over 2–3 months. Depression improvements may take 4–8 weeks to become noticeable. Gut microbiota shifts occur within 2–4 weeks.

**Important caveat for n=1:** You cannot prove causation from a single experiment without a control period. The strongest design would be an **ABAB crossover**: 3 months on almond-LCD, 3 months on your usual diet, then repeat. This controls for time trends and placebo effects. However, 12 months is a long commitment. A simpler approach: do 3 months on almond-LCD, then 3 months on low-fat diet (or vice versa), and compare the two periods. Even then, order effects (you might improve simply because you're paying attention to your health) can't be ruled out.

**Bottom line:** This paper suggests that combining almonds with carbohydrate restriction may improve both depression and blood sugar in type 2 diabetes, possibly through gut bacteria and GLP-1. The effect is plausible but the evidence is weak (small study, no blinding, no effect sizes reported). For a self-experiment, it's worth testing — but track everything, control for confounds, and don't expect miracles. A 0.5% drop in HbA1c and a 5-point drop in PHQ-9 would be a solid win.