Sweet Potato Glycemic Index

Table of Contents

1. Glycemic Index vs Glycemic Load: The Distinction That Matters
2. Measured Glycemic Index Values for Sweet Potato
3. Sweet Potato vs White Potato: The Measured Values
4. How Cooking Method Changes the GI
5. The Cool-and-Reheat Resistant Starch Effect
6. Mechanisms: Why Sweet Potato Is Slower
7. The White-Skinned Sweet Potato Extract Diabetes Trials
8. Practical Portions for Diabetics and Insulin-Resistant Eaters
9. Combining with Fat, Protein, and Vinegar
10. Key Research Papers
11. Connections
12. Featured Videos

Glycemic Index vs Glycemic Load: The Distinction That Matters

Before walking through the numbers, the distinction between glycemic index (GI) and glycemic load (GL) matters because they answer different questions and are easy to confuse.

Glycemic index is a property of a food itself: when you feed a standardized 50 g portion of available carbohydrate from a specific food to a human subject, how much does their blood glucose rise relative to a reference (usually pure glucose at GI 100)? The reference question is "per gram of carbohydrate, how fast does this food spike my blood sugar?" GI is independent of portion size.

Glycemic load is a property of a serving: GL = (GI × grams of carbohydrate per serving) / 100. The reference question is "how much will this actual portion I am about to eat raise my blood sugar?" GL combines the food's intrinsic glycemic property with the realistic portion size.

Sweet potato is in the low-GI category (under 55) and in the low-to-moderate GL category (under 10 to 15 per serving). A 200 g cooked sweet potato contains approximately 28 g of available carbohydrate, so the glycemic load of one medium boiled sweet potato is approximately (44 × 28) / 100 = 12 — a moderate-to-low load that is well-tolerated by most people with diabetes when paired with protein and fat.

For comparison, a 200 g baked white potato (also ~28 g carbohydrate) has a glycemic load of (78 × 28) / 100 = 22 — nearly twice the postprandial glucose impact for the identical portion.

Back to Table of Contents

Measured Glycemic Index Values for Sweet Potato

The published GI values for sweet potato vary by cultivar, ripeness, cooking method, country of origin, and measurement protocol. The Atkinson, Foster-Powell, and Brand-Miller international GI tables (2008 edition, and the 2021 edition with over 4,000 entries) list the following representative values:

• Boiled sweet potato (peeled, 30 min): GI 44 (Australia)
• Boiled sweet potato (peeled, 25 min): GI 46 (USA)
• Baked sweet potato (skin on, 45 min at 180°C): GI 64 (Australia)
• Roasted sweet potato (60 min at 200°C): GI 82 (USA, one study)
• Sweet potato chips/fries: GI 76 (variable)
• Sweet potato pureé: GI 54 (UK)
• Sweet potato Beauregard variety: GI 62 (USA)
• Sweet potato Hayman White variety: GI 39 (USA)

Measured values for baked sweet potato vary widely between studies. In a Jamaican study of ten sweet potato varieties eaten by ten healthy adults, boiled tubers had a GI of 41–50, while baked tubers (82–94) and roasted tubers (79–93) were in the high range — well above the single baked value of 64 listed above. The authors found that the GI varied significantly with the method of preparation and to a lesser extent between varieties.

Three practical takeaways from these numbers:

1. Boiling produces the lowest GI. The water-immersion cooking limits starch gelatinization, preserves some resistant starch, and avoids the Maillard browning that occurs with dry-heat cooking.
2. Baking and roasting roughly double the GI compared to boiling, because the higher dry-heat temperatures more fully gelatinize the starch, making it more accessible to amylase digestion.
3. Cultivar effects are real but smaller than cooking effects. In the ten-variety Jamaican study, the cooking method changed the GI more than the choice of variety did. Hayman White at GI 39 versus Beauregard at GI 62 is a notable difference, but the boiled-versus-baked difference within a single cultivar is larger.

Back to Table of Contents

Sweet Potato vs White Potato: The Measured Values

The most asked question about sweet potato is "is it really better than regular potato for blood sugar?" Boiled sweet potato measures lower than most hot white potatoes, and the size of the gap depends on the cooking method, the variety and the serving temperature.

The cleanest head-to-head is boiled-versus-boiled. Boiled sweet potato (GI 44) versus boiled Russet potato (GI 70) is a 26-point gap. For baked-versus-baked, the table value for baked sweet potato (GI 64) versus baked Russet (GI 78) is a 14-point gap, but the baked values of 82–94 measured across ten Jamaican varieties are at or above the baked Russet figure. These are single values: the 2008 international tables average 63 for boiled sweet potato and 78 for boiled potato, and the same boiled red potato measured 89 hot and 56 cold in a North American study (Fernandes 2005; more on the Potatoes page). The general pattern: boiled sweet potato measures lower than hot boiled white potato, while after baking the gap shrinks and may disappear.

For a person with type 2 diabetes consuming 30 g of available carbohydrate from a starchy side dish, the swap from baked Russet to boiled sweet potato will reduce their two-hour postprandial glucose peak by roughly 15–30 mg/dL on average — not a dramatic effect, but a meaningful one that compounds across daily meals over months and years.

Back to Table of Contents

How Cooking Method Changes the GI

The same tuber prepared by different methods produces dramatically different glycemic responses. Starting from highest GI and working down:

1. Roasted at high temperature (200°C+) for 60+ minutes — GI 75–85. Maximum starch gelatinization, Maillard browning of surface starches, complete cell-wall breakdown. Almost as glycemic as baked white potato.
2. Microwave-baked, whole, skin on — GI ~70. Comparable to oven-baked but slightly less surface caramelization.
3. Oven-baked, skin on, 45 min at 180°C — GI ~64 in the international tables, but 82–94 across ten varieties in the Jamaican study. The classic American preparation.
4. Sweet potato fries (oven, with olive oil) — GI ~62 with fat slowing absorption; fat does not change the food's intrinsic GI but does slow gastric emptying.
5. Mashed (boiled then crushed) — GI ~55. The mashing increases surface area for amylase but the underlying boiled-starch profile remains.
6. Boiled, peeled, in chunks, 30 minutes — GI ~46. The reference low-GI preparation.
7. Boiled, then cooled and reheated — GI ~40 (see next section on resistant starch).
8. Steamed, skin on — GI ~44. Comparable to boiled, slightly more nutrient retention.

The biggest single change is the boiling-versus-roasting spread of roughly 30 GI points on the same tuber. For a person managing blood sugar, that is more impact than switching from sweet potato to a different food. For high-GI preparations such as roasted sweet potatoes at a holiday meal, protein, fat, and acid in the same meal flatten the curve (see below), and boiling or steaming gives the lowest values.

Back to Table of Contents

The Cool-and-Reheat Resistant Starch Effect

One of the most underappreciated metabolic interventions is starch retrogradation. When cooked starch is cooled below approximately 5°C, the gelatinized starch chains begin to re-crystallize into a structure that human amylase cannot digest — functionally, it becomes fiber. This fraction is called type 3 resistant starch (RS3).

The sequence usually described:

1. The sweet potato is boiled or baked as usual.
2. The cooked sweet potato is refrigerated for at least 12 hours (overnight).
3. It is eaten cold (in a salad, for example), or reheated gently.

The reheated sweet potato now contains substantially more resistant starch than the freshly cooked tuber. Estimates suggest 5–10 percentage points of the total starch is converted to RS3 by overnight refrigeration. The metabolic consequences:

• Lower glycemic response — some of the carbohydrate is no longer rapidly absorbed
• Lower effective calorie content — resistant starch provides ~2 kcal/g (fermentation byproducts) versus 4 kcal/g for digested starch
• Increased colonic short-chain fatty acid production — resistant starch is the favored substrate for butyrate-producing colonic bacteria
• Improved insulin sensitivity at the next meal (the "second-meal effect")

This is the same trick described for white rice, pasta, and potato, though potato shows limits: cooking the day before and reheating lowered the glucose response for Russet potatoes but not for boiled white potatoes (Fernandes 2005). It is one of the few free-lunch interventions in nutrition — the same food, the same total calorie intake, but a measurably better metabolic profile achieved entirely by a refrigeration step.

Back to Table of Contents

Mechanisms: Why Sweet Potato Is Slower

Several mechanisms combine to produce the lower glycemic index of sweet potato relative to white potato.

1. Higher amylose-to-amylopectin ratio. Plant starches consist of two glucose polymers: linear amylose and branched amylopectin. Amylose is digested more slowly because its long unbranched chains are more difficult for amylase to access. White potato starch is roughly 80% amylopectin and 20% amylose — weighted toward fast digestion. Sweet potato starch is closer to 70% amylopectin and 30% amylose, shifting the balance toward slower digestion.

2. Dietary fiber content. Sweet potato contains 3–4 g of dietary fiber per 100 g (cooked), compared to 1.5–2.5 g for white potato. Fiber slows gastric emptying, slows enzymatic access to starch granules, and contributes some non-digestible carbohydrate that does not raise blood glucose.

3. Intact cell wall structure. The sweet potato's tuber tissue has thicker, more lignified cell walls than the white potato. Even after cooking, some of the starch granules remain encapsulated within partially intact cell walls, physically delaying amylase access. White potato cell walls largely disintegrate under cooking heat.

4. Polyphenol alpha-amylase inhibition. Sweet potato contains chlorogenic acid, caffeic acid, and other phenolic compounds that competitively inhibit pancreatic alpha-amylase and intestinal brush-border alpha-glucosidase. The effect is small but measurable — comparable in magnitude to a low dose of the diabetic drug acarbose. White potato has substantially less polyphenol content (its polyphenols are concentrated in the skin, which is often discarded).

5. Resistant starch baseline. Even freshly cooked, sweet potato contains 2–4% resistant starch versus 1–2% for white potato. The retrogradation effect described in the previous section adds to this baseline.

These mechanisms are additive, not redundant. Each contributes a few GI points, and the combined effect of all five mechanisms is offered as the explanation for the lower values of boiled sweet potato. The gap is not fixed: the single values in the table above give 26 points when both are boiled and 14 points or less when both are baked.

Back to Table of Contents

The White-Skinned Sweet Potato Extract Diabetes Trials

An often-overlooked set of clinical trials used an extract of white-skinned sweet potato (Ipomoea batatas), sold in Japan as a nutraceutical. Sweet potato is used in traditional medicine for type 2 diabetes. The Ludvik et al. studies (2002, 2003, 2004, 2008), all from the same research group in Vienna, randomized people with type 2 diabetes treated by diet alone to the extract or placebo:

• 2003 (6 weeks): 18 men were given placebo, 2 g/day or 4 g/day of the extract. At 4 g/day insulin sensitivity and glucose tolerance improved, without changes in body weight or insulin secretion.
• 2004 (12 weeks): 61 people were given 4 g/day of the extract (30 people) or placebo (31 people). Hemoglobin A1c (HbA1c) fell from 7.21% to 6.68% on the extract and stayed unchanged on placebo (7.04% to 7.10%). Fasting blood glucose fell from 143.7 to 128.5 mg/dL on the extract. Mean cholesterol at the end of the trial was lower on the extract (214.6 mg/dL) than on placebo (248.7 mg/dL). The extract was well tolerated, with no significant adverse effects.
• 2008 (5 months): 27 people took 4 g/day of the extract and 34 took placebo. On the extract, insulin sensitivity (measured as oral glucose insulin sensitivity, OGIS) rose from 293 to 321 ml/m²/min, adiponectin rose, fibrinogen fell, and HbA1c went from 6.46% to 6.25%; on placebo, HbA1c rose from 6.25% to 6.50%.

What the Cochrane review found. A 2013 Cochrane review found only three small randomized trials of sweet potato preparations (140 participants in total, lasting six weeks to five months), all by the same researcher, with an unclear or high risk of bias. Pooling two trials (122 participants), HbA1c was about 0.3 percentage points lower with 4 g/day than with placebo (95% confidence interval -0.6 to -0.04). No serious adverse effects were reported. The reviewers concluded that “there is insufficient evidence about the use of sweet potato for type 2 diabetes mellitus.”

The mechanism for the extract's glycemic effect is not fully established. Hypotheses include enhanced insulin sensitivity at the muscle level, alpha-glucosidase inhibition, anti-inflammatory effects, and possibly direct insulin-mimetic activity from the storage protein sporamin.

The clinical reach of this finding is limited — the extract comes from a white-skinned sweet potato, not the orange sweet potato readily available in U.S. grocery stores, and the trials used a concentrated extract rather than whole food. But the result is suggestive that the metabolic benefits of sweet potato consumption go beyond the passive low-GI property to include some active insulin-sensitizing effect at the tissue level. This is consistent with epidemiologic data showing that populations with high sweet-potato consumption (e.g. traditional Okinawan diet) have very low diabetes prevalence.

Back to Table of Contents

Practical Portions for Diabetics and Insulin-Resistant Eaters

For people with type 2 diabetes, pre-diabetes, insulin resistance, or polycystic ovary syndrome-related glucose dysregulation, sweet potato is one of the better starchy carbohydrate options, but portion control still matters.

Common serving: 150–200 g cooked (one medium-small tuber), which is ~22–28 g of available carbohydrate. For a diabetic eating a 30–45 g carbohydrate target per meal, this is a substantial single carbohydrate source; a second starchy carbohydrate (bread, rice, pasta) at the same meal adds to that total.

Larger serving for active or non-diabetic eaters: 250–300 g cooked (one large tuber or two small), ~35–42 g carbohydrate. Often eaten as a post-workout meal or by non-diabetic readers who tolerate carbohydrate well.

Frequency: The fiber, micronutrient density, and low glycemic load of boiled sweet potato are why it is commonly kept on diabetes-focused diets most days of the week rather than restricted.

Timing: Many readers report better postprandial response when sweet potato is consumed at lunch or dinner with substantial protein and fat, rather than as a stand-alone breakfast or snack carbohydrate. The mixed meal pattern flattens the glucose curve.

Higher-sugar and fried forms: Sweet potato pie (added sugar), candied sweet potatoes with marshmallow topping (added sugar plus high-glycemic refined sugar), sweet potato chips (deep-fried in seed oils, high glycemic and inflammatory), commercial sweet potato fries (typically deep-fried in vegetable oil with added starch coatings).

Back to Table of Contents

Combining with Fat, Protein, and Vinegar

Three meal-construction principles flatten the glycemic response to any starchy carbohydrate, including sweet potato:

Add fat. Fat slows gastric emptying. Sweet potato eaten with butter, olive oil, coconut oil, avocado, tahini, or alongside a fatty protein (salmon, eggs, beef) produces a slower glucose rise than sweet potato eaten alone. Eating it with fat is also linked with greater absorption of its fat-soluble carotenoids (see the Beta-Carotene deep-dive).

Add protein. Protein co-ingestion stimulates a small insulin response that helps clear circulating glucose, and also slows gastric emptying. Sweet potato paired with eggs, fish, meat, beans, or Greek yogurt is metabolically gentler than sweet potato as a stand-alone starch.

Add acid. Vinegar (acetic acid) is the best-studied acid for blunting postprandial glucose. A few teaspoons of apple cider vinegar or balsamic vinegar in a dressing on sweet-potato salad is the usual way it is added. A few teaspoons of apple cider vinegar or balsamic vinegar in a dressing on sweet-potato salad, or a squeeze of lemon, reduces the glycemic peak by 20–30% in repeated clinical trials. A 2017 meta-analysis of controlled trials found that vinegar taken with a meal lowered the blood-glucose response (the area under the glucose curve; standardized mean difference -0.60, 95% confidence interval -1.08 to -0.11) and the insulin response (standardized mean difference -1.30, 95% CI -1.98 to -0.62) compared with control meals. Those pooled trials tested vinegar; lemon juice is a separate question. In a small Swedish study of ten healthy adults, vinegar added to a starchy meal lowered the glucose and insulin responses, probably by slowing how fast the stomach empties; direct inhibition of disaccharidase activity in the small intestine is another proposed mechanism.

A practical example: 200 g boiled sweet potato cubes, 100 g grilled chicken, two tablespoons olive oil, one tablespoon apple cider vinegar, one cup of mixed greens. This produces a meaningfully smaller postprandial glucose excursion than 200 g baked Russet potato, butter, and a similar protein portion.

Back to Table of Contents

Key Research Papers

1. Atkinson FS, Foster-Powell K, Brand-Miller JC — International tables of glycemic index and glycemic load values: 2008. — PubMed PMID: 18835944
2. Atkinson FS, Brand-Miller JC, Foster-Powell K, Buyken AE, Goletzke J — International tables of glycemic index and glycemic load values 2021: a systematic review. Am J Clin Nutr 2021;114(5):1625-1632. — PubMed PMID: 34258626
3. Ludvik BH, Mahdjoobian K, Waldhaeusl W, et al. — The effect of Ipomoea batatas (Caiapo) on glucose metabolism and serum cholesterol in patients with type 2 diabetes: a randomized study. Diabetes Care 2002;25(1):239-40. — PubMed PMID: 11772921
4. Ludvik B, Neuffer B, Pacini G — Efficacy of Ipomoea batatas (Caiapo) on diabetes control in type 2 diabetic subjects treated with diet. Diabetes Care 2004;27(2):436-40. — PubMed PMID: 14747225
5. Ludvik B, Hanefeld M, Pacini G — Improved metabolic control by Ipomoea batatas (Caiapo) is associated with increased adiponectin and decreased fibrinogen levels in type 2 diabetic subjects. Diabetes Obes Metab 2008;10(7):586-92. — PubMed PMID: 17645559
6. Ooi CP, Loke SC — Sweet potato for type 2 diabetes mellitus. Cochrane Database Syst Rev 2013;(9):CD009128. — PubMed PMID: 24000051
7. Ludvik B et al. — Mode of action of Ipomoea batatas (Caiapo) in type 2 diabetic patients. — PubMed PMID: 12870164
8. Bahado-Singh PS, Riley CK, Wheatley AO, Lowe HI — Relationship between processing method and the glycemic indices of ten sweet potato (Ipomoea batatas) cultivars commonly consumed in Jamaica. J Nutr Metab 2011;2011:584832. — PubMed PMID: 22132322
9. Englyst HN, Kingman SM, Cummings JH — Classification and measurement of nutritionally important starch fractions. Eur J Clin Nutr 1992;46 Suppl 2:S33-50. — PubMed PMID: 1330528
10. Sajilata MG, Singhal RS, Kulkarni PR — Resistant starch — a review. Compr Rev Food Sci Food Saf 2006;5(1):1-17. — PubMed PMID: 33412740
11. Johnston KL, Thomas EL, Bell JD, et al. — Resistant starch improves insulin sensitivity in metabolic syndrome. Diabet Med 2010;27(4):391-7. — PubMed PMID: 20536509
12. Robertson MD et al. — Insulin-sensitizing effects on muscle and adipose tissue after dietary fiber intake in men and women with metabolic syndrome. J Clin Endocrinol Metab 2012;97(9):3326-32. — PubMed PMID: 22745235
13. Liljeberg H, Björck I — Delayed gastric emptying rate may explain improved glycaemia in healthy subjects to a starchy meal with added vinegar. Eur J Clin Nutr 1998;52(5):368-71. — PubMed PMID: 9630389
14. Shishehbor F, Mansoori A, Shirani F — Vinegar consumption can attenuate postprandial glucose and insulin responses; a systematic review and meta-analysis of clinical trials. Diabetes Res Clin Pract 2017;127:1-9. — PubMed PMID: 28292654
15. Johnston CS et al. — Vinegar improves insulin sensitivity to a high-carbohydrate meal in subjects with insulin resistance or type 2 diabetes. — PubMed PMID: 14694010
16. Fernandes G, Velangi A, Wolever TM — Glycemic index of potatoes commonly consumed in North America. J Am Diet Assoc 2005;105(4):557-62. — PubMed PMID: 15800557
17. Henry CJ, Lightowler HJ, Strik CM, Storey M — Glycaemic index values for commercially available potatoes in Great Britain. Br J Nutr 2005;94(6):917-21. — PubMed PMID: 16351768

Back to Table of Contents

Connections

• All Food
• Sweet Potato Benefits (Hub)
• Sweet Potatoes (Main Page)
• Beta-Carotene and Vision
• Purple vs Orange Cultivars
• Skin Nutrient Density
• Type 2 Diabetes
• Insulin Resistance
• Metabolic Syndrome
• Hemoglobin A1c
• Fasting Glucose
• Lentils (Low-GI Carbohydrate)
• Apple Cider Vinegar
• History
• Nutrients

Back to Table of Contents

×

Featured Videos