HIIT (High-Intensity Interval Training)

What it is

High-intensity interval training (HIIT) alternates short bursts of hard exercise with rest or easier exercise [1]. Protocols vary widely: short intervals are 30 seconds or less, and long intervals are two minutes or more [2]. Sprint interval training is one form, and some studies used work periods under 10 seconds [3]. Researchers study HIIT partly because it needs less time than moderate-intensity continuous training (MICT), which is steady exercise at a moderate pace [1].

How it might work

Proposed mechanism for fitness: in a review of trials, HIIT raised maximal stroke volume (blood pumped per heartbeat) by 14.1% and maximal cardiac output by 12.6%, which suggests the heart adapts [4]. Proposed mechanism for blood pressure: in a 2026 trial in adults with hypertension, blood pressure drops were linked to falls in the gut-derived compound TMAO, but this is an association, not proof of cause [5]. Proposed mechanism for blood sugar: in a small trial in type 2 diabetes, HIIT reduced how fast glucose from a meal entered the bloodstream [6].

What the research says

Fitness has the strongest support. A 2026 Cochrane review of 58 trials with 2,075 healthy, sedentary adults found that HIIT raised VO2max, a measure of aerobic fitness, by about 6 mL per kilogram per minute more than no exercise [1]. Other meta-analyses also find that HIIT improves VO2max, but the size of the gain depends on the protocol and the group studied [2].

Against moderate continuous training, the Cochrane review found a small extra VO2max gain of about 1.4 mL/min/kg, with low certainty [1]. The extra gain appeared only with longer intervals, higher training volume and longer programmes [2]. In a meta-analysis of adults with overweight or obesity, moderate training improved fitness more than sprint intervals did [7].

Time is the main reason to choose HIIT. In an 11-week trial of 29 adults with type 2 diabetes, HIIT gave similar or better fitness and glucose results than steady cycling, with about 45% less training volume [6].

Blood pressure: in a 2024 meta-analysis of 23 trials with 1,374 adults with metabolic syndrome, HIIT lowered systolic blood pressure by about 6 mmHg more than no exercise [8]. A 12-week trial of 60 adults with hypertension found office blood pressure fell by 6.9 and 3.1 mmHg (systolic and diastolic) more than in controls [5]. The evidence is less certain than these numbers suggest. The Cochrane review found a similar average drop, but its range includes no effect, so its certainty is very low [1]. A network meta-analysis of 270 trials ranked HIIT lowest of five exercise types for systolic pressure, though all five reduced it [9].

Blood sugar in type 2 diabetes: an umbrella review of 10 systematic reviews found that HIIT lowered HbA1c, a measure of average blood sugar, by 0.39% to 0.83% more than no exercise [10]. Compared with moderate training, the difference was smaller, 0.07% to 0.37% [10]. A meta-analysis of 29 trials found that long, high-volume, long-term HIIT lowered HbA1c more than no exercise, with no difference from moderate training for glucose control [11]. Not every trial agrees: a one-year randomised trial of 80 adults found no HbA1c effect for HIIT combined with resistance training [12].

Waist size and body fat: the Cochrane review found waist circumference fell by about 3.6 cm more than with no exercise, and it rates that finding as high certainty [1]. A 2024 meta-analysis of adults with metabolic syndrome found a 4.1 cm reduction against controls [8]. In children with overweight or obesity, low-volume HIIT lowered body fat percentage more than no exercise and about the same as moderate training [13]. In adults with overweight or obesity, body fat did not differ between HIIT and moderate training [7].

Enjoyment: eight studies of single sessions found that HIIT was rated as more enjoyable than moderate exercise on one scale, with a small effect, while a second scale showed no significant difference [14]. In 42 adults with overweight or obesity, enjoyment rose over three weeks of HIIT [15]. The reviewers note that longer studies are needed to show whether people keep it up [14].

Doses used in studies

Protocols that improved fitness ranged from short intervals (30 seconds or less, up to 5 minutes of hard work, up to 4 weeks) to longer ones (2 minutes or more, 15 minutes or more of hard work, 4 to 12 weeks). The longer ones gave larger gains [2].

In a 12-week trial in adults with hypertension, participants did 10 to 15 one-minute cycling intervals at 85% to 90% of heart rate reserve, with one minute of easier recovery, three times a week [5].

In an 11-week trial in type 2 diabetes, participants did 10 one-minute efforts at 95% of peak workload with one minute of recovery, three sessions a week. A comparison group did 40 minutes of cycling at 50% of peak workload [6].

Safety and interactions

Side effects: in supervised cardiac rehabilitation, a review of 23 studies with 1,117 people recorded one major heart event linked to HIIT sessions, about one per 17,000 sessions [16]. The other reported problems were mostly muscle and joint complaints [16].

Upper limits: none of the reviews found here reported a toxic dose or a safe upper limit for training volume [1].

Medicines: our searches found no studies of HIIT combined with medicines.

Pregnancy and breastfeeding: our searches found no trials of HIIT in pregnancy or breastfeeding.

Heart disease and blood pressure: the cardiac rehabilitation review found the risk of serious events low when HIIT was supervised [16]. In adults with congenital heart disease, three studies with 87 people reported no serious adverse events, but the authors call the data too limited to rely on [17]. A review in people with hypertension found HIIT appeared safe only when supervised, and only in people with stage 1 hypertension and no other risk factors [18].

Kidney, liver and bleeding disorders: our searches found no data on these groups.

Unsupervised training: all trials in the Cochrane review were supervised, and the authors say the safety of unsupervised HIIT needs study [1].

Caveats and open questions

Most trials compared HIIT with no exercise. Against moderate training, results were usually similar or small [1].

Many trials are small. The Cochrane review notes that its included studies all had relatively few participants [1].

Blood pressure and blood sugar evidence is less certain than fitness evidence. The diabetes umbrella review rated most of the reviews it included as moderate to low quality [10].

Adherence is untested. The enjoyment review found only single sessions and called for longer studies on adherence [14].

Some trials focus on special groups, such as children with overweight or obesity [13] or people with multiple sclerosis [19]. Results may not apply directly to other adults.