High-Intensity Interval Training and Moderate-Intensity Continuous Training Affect Running Economy in Endurance Runners: A Systematic Review and Meta-Analysis of Randomized Controlled Trials.

This meta-analysis compared two common endurance training methods head-to-head:

**HIIT (High-Intensity Interval Training):** Repeated short bursts of hard running (typically 30 seconds to 4 minutes at 90–100% of maximal heart rate or at or above lactate threshold) separated by recovery periods (walking or jogging). Total session time is usually 15–30 minutes.

**MICT (Moderate-Intensity Continuous Training):** Steady-state running at a moderate pace (typically 60–75% of maximal heart rate, below lactate threshold) for 30–60 minutes continuously.

The primary outcome was **running economy (RE)** — how much oxygen you consume (ml·kg⁻¹·min⁻¹) at a given submaximal running speed. A runner with better economy uses less oxygen at the same pace, meaning they can run faster or longer with the same effort.

Secondary outcomes were:

**VO₂max** (maximal oxygen uptake, ml·kg⁻¹·min⁻¹) — the ceiling of your aerobic engine

**Blood lactate concentration** (mmol/L) — a marker of metabolic stress and fatigue

The researchers also split running economy into three intensity zones:

**Zone 1 (Z1):** Below lactate threshold (easy conversational pace)

**Zone 2 (Z2):** At or near lactate threshold (moderately hard, sustainable for ~1 hour)

**Zone 3 (Z3):** Above lactate threshold (hard, sustainable for <30 minutes)

**Total sample:** 18 randomized controlled trials (RCTs) were included in the meta-analysis

**Total participants:** 388 endurance runners (exact breakdown by sex not consistently reported across studies)

**Population:** Trained endurance runners (competitive or recreational), including distance runners, middle-distance runners, and cross-country runners

**Age range:** Approximately 18–45 years (varied by individual study)

**Training status:** All participants were already engaged in regular endurance running training (not sedentary beginners)

**Setting:** Laboratory-based treadmill testing with supervised training programs

**Exclusion criteria:** Studies with training interventions lasting less than 4 weeks were excluded; non-randomized trials were excluded; studies not published in English were excluded

**Running economy (RE):** Measured as steady-state oxygen consumption (VO₂, ml·kg⁻¹·min⁻¹) or energy cost (kcal·kg⁻¹·km⁻¹) during submaximal treadmill running at a fixed speed. Participants ran on a treadmill while wearing a metabolic cart (portable or stationary gas analyzer) that measured inspired and expired gases breath-by-breath. Measurements were taken after 3–6 minutes of steady-state running at each speed.

**VO₂max:** Measured using a graded exercise test on a treadmill (typically starting at a moderate speed and increasing incline or speed every 1–3 minutes until volitional exhaustion). The highest 30-second average of oxygen consumption was recorded as VO₂max.

**Blood lactate:** Measured via finger-prick or earlobe blood samples taken immediately after submaximal running stages or at the end of exercise. Analyzed using portable lactate analyzers (e.g., Lactate Pro, Biosen) or enzymatic methods.

**Intensity zones for RE:** Studies reported RE at specific running speeds corresponding to:

- Z1: 60–75% of VO₂max or below lactate threshold (LT)

- Z2: 75–85% of VO₂max or at LT

- Z3: 85–95% of VO₂max or above LT

**Study design:** Systematic review and meta-analysis of randomized controlled trials (RCTs). The review followed PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines and was registered with INPLASY (registration number: INPLASY2024110120).

**Search strategy:** Four databases (PubMed, Embase, Scopus, Web of Science) were searched from their inception to March 2024. Search terms included combinations of running-related terms (running, jogging, marathon running, distance runners) with running economy terms (energy metabolism, energy cost, metabolic cost) and training terms (interval training, sprint interval training, intensity training). Only English-language RCTs were included.

**Inclusion criteria (PICOS):**

**Participants:** Endurance runners (trained, not sedentary)

**Intervention:** Interval training (HIIT)

**Comparator:** Continuous training (MICT)

**Outcomes:** Running economy or energy consumption

**Study design:** Peer-reviewed RCTs

**Duration:** Minimum 4 weeks of training intervention

**Exclusion criteria:** Studies without specified intervention type, non-RCTs, training interventions shorter than 4 weeks, non-interval training, review articles.

**Data extraction:** Two independent reviewers screened titles/abstracts, then full texts. A third reviewer resolved disagreements. Data extracted included: author, year, country, participant characteristics, VO₂max, training protocols, RE measurements, and lactate values. Post-intervention means and standard deviations were used for effect size calculations. When SDs were missing, they were calculated from SEs, confidence intervals, t-values, or p-values.

**Risk of bias assessment:** Two reviewers independently assessed bias using the Cochrane Risk of Bias Tool, evaluating: random sequence generation, allocation concealment, blinding of participants and personnel (noted as impossible to blind participants to training type), blinding of outcome assessment, incomplete outcome data, selective reporting, and other biases. Performance bias (blinding of participants) was excluded because it was not feasible. Sensitivity analyses excluded high-risk studies to check robustness.

**Statistical analysis:** Meta-analysis was performed using RevMan 5.3.5. For continuous outcomes:

**Running economy:** Standardized mean difference (SMD) was used because different studies measured RE in different units (ml·kg⁻¹·min⁻¹ vs. kcal·kg⁻¹·km⁻¹). SMD = 0.44 means HIIT improved RE by 0.44 standard deviations more than MICT.

**VO₂max:** Mean difference (MD) was used (same units across studies). MD = 2.48 ml·kg⁻¹·min⁻¹.

**Blood lactate:** Mean difference (MD) was used. MD = −0.15 mmol/L.

Subgroup analyses were conducted by intensity zone (Z1, Z2, Z3). A funnel plot was used to assess publication bias for significant results (p < 0.05).

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

**Can prove:**

That HIIT produces statistically different effects on RE, VO₂max, and blood lactate compared to MICT in trained endurance runners

The direction and magnitude of these differences (HIIT better for RE and lactate; MICT better for VO₂max)

That effects vary by training intensity zone (HIIT works best for RE at or below lactate threshold)

**Cannot prove:**

Causality at the individual level (meta-analyses aggregate group means, not individual responses)

Why these differences occur (mechanisms are inferred from individual studies, not tested directly)

Long-term effects beyond the study durations (most studies were 4–12 weeks)

Effects in untrained populations, non-runners, or elite athletes specifically (participants were "trained endurance runners" but not necessarily elite)

Optimal specific HIIT or MICT protocols (the meta-analysis combined many different protocols)

Whether combining HIIT and MICT is better than either alone (no combined training group was analyzed)

**Methodological weaknesses:**

**Blinding impossible:** Participants knew which training they were doing, which could affect motivation, effort, and adherence

**Protocol heterogeneity:** HIIT protocols varied widely (work:rest ratios, interval durations, intensities). MICT also varied in duration and intensity. This makes it hard to pinpoint which specific protocol works best.

**Language bias:** Only English-language studies were included, potentially excluding relevant non-English research

**Publication bias possible:** Studies with null or negative results are less likely to be published

**Small number of studies per subgroup:** Some subgroup analyses (especially Z3) had very few studies, reducing statistical power

**No assessment of training volume matching:** It's unclear whether total work or total energy expenditure was matched between HIIT and MICT groups across studies

**Short intervention durations:** Most studies were 4–12 weeks; longer-term adaptations may differ

**Primary outcome — Running economy (RE):**

HIIT significantly improved RE compared to MICT overall: **SMD = 0.44, 95% CI [0.15, 0.72], Z = 3.01, p < 0.05**

- This is a **moderate effect size** (Cohen's d: 0.2 = small, 0.5 = moderate, 0.8 = large)

Subgroup analysis by intensity zone:

- **Zone 1 (below lactate threshold):** HIIT superior to MICT (SMD = 0.51, 95% CI [0.18, 0.84], p < 0.05) — **moderate effect**

- **Zone 2 (at lactate threshold):** HIIT superior to MICT (SMD = 0.47, 95% CI [0.12, 0.82], p < 0.05) — **moderate effect**

- **Zone 3 (above lactate threshold):** No significant difference between HIIT and MICT (SMD = 0.23, 95% CI [−0.18, 0.64], p > 0.05) — **small, non-significant effect**

**Secondary outcome — VO₂max:**

MICT produced significantly greater improvements in VO₂max compared to HIIT: **MD = 2.48 ml·kg⁻¹·min⁻¹, 95% CI [1.61, 3.34], Z = 5.60, p < 0.05**

- This means MICT increased VO₂max by about 2.5 ml·kg⁻¹·min⁻¹ more than HIIT, on average

- For context: a typical trained runner has a VO₂max of 45–60 ml·kg⁻¹·min⁻¹; a 2.5 unit increase represents roughly a 4–5% improvement

**Secondary outcome — Blood lactate:**

HIIT was more effective at reducing blood lactate levels compared to MICT: **MD = −0.15 mmol/L, 95% CI [−0.28, −0.02], Z = 2.20, p < 0.05**

- This is a **small effect** — HIIT reduced lactate by 0.15 mmol/L more than MICT at the same running speed

- For context: resting lactate is ~1 mmol/L; during moderate running it rises to 2–4 mmol/L; lactate threshold is typically around 4 mmol/L

**Risk of bias assessment:**

Most studies had **low or unclear risk of bias** for random sequence generation and allocation concealment

**High risk of performance bias** was noted (participants couldn't be blinded to training type)

**Low risk** for incomplete outcome data and selective reporting in most studies

Sensitivity analyses excluding high-risk studies showed the overall results remained robust

**Publication bias:**

Funnel plot analysis was performed for significant results (p < 0.05) and suggested no major publication bias, though the small number of studies limits this assessment

**Running economy:** The SMD of 0.44 means that if you randomly picked a runner from the HIIT group and a runner from the MICT group, the HIIT runner would have better running economy about 63% of the time (based on the probability of superiority for a Cohen's d of 0.44). In practical terms, this translates to roughly a **1–3% improvement in oxygen consumption at a given speed** — meaning at 12 km/h, a runner might use 1–2 ml·kg⁻¹·min⁻¹ less oxygen after HIIT compared to MICT. For a marathon runner, a 1% improvement in running economy can translate to a 2–3 minute faster marathon time.

**VO₂max:** The 2.48 ml·kg⁻¹·min⁻¹ advantage for MICT is meaningful. For a runner with a VO₂max of 50 ml·kg⁻¹·min⁻¹, this represents a ~5% greater improvement with MICT. This is roughly equivalent to the difference between a good club runner and an average recreational runner.

**Blood lactate:** The 0.15 mmol/L reduction with HIIT is small but meaningful at the margins. At lactate threshold (typically 4 mmol/L), this represents a ~4% reduction. This means HIIT runners can run slightly faster before accumulating lactate, which delays fatigue.

**Zone-specific effects:** The finding that HIIT works best for RE at or below lactate threshold (Z1 and Z2) is practically important. This suggests HIIT improves the efficiency of your "aerobic engine" at the paces you actually race (marathon pace is typically at or just below lactate threshold), not just at very high intensities.

**What the authors acknowledge:**

Performance bias was unavoidable (participants knew their training group)

Only English-language studies were included

Training protocols varied considerably across studies

The number of studies in some subgroup analyses was small

Publication bias could not be fully ruled out

**What a critical reader would note:**

**Short training durations:** Most included studies were 4–12 weeks. Longer training periods (6+ months) might show different patterns — MICT might catch up on RE, or HIIT might plateau.

**No dose-response analysis:** The meta-analysis didn't examine whether more HIIT sessions per week or longer HIIT intervals produce larger effects. You can't tell if "more is better."

**Training volume not matched:** It's unclear whether HIIT and MICT groups did the same total amount of work. HIIT sessions are typically shorter (15–30 minutes vs. 30–60 minutes for MICT). If total work was lower in HIIT, the RE advantage is even more impressive. If total work was matched, the time-saving benefit of HIIT is reduced.

**Population limits:** All participants were already trained runners. Results may not apply to beginners, older adults, or elite athletes. Trained runners have less room for improvement, so effects might be larger in untrained populations.

**No long-term follow-up:** Studies measured effects immediately after the training period. We don't know if RE improvements from HIIT persist after stopping training, or if they require ongoing HIIT to maintain.

**Individual variability:** Meta-analyses report averages. Some runners may respond much better to HIIT, others to MICT. The "best" approach may depend on individual genetics, training history, and running biomechanics.

**No control for running form changes:** RE improvements could come from biomechanical changes (better running form) rather than physiological adaptations. The meta-analysis didn't assess running mechanics.

**Industry funding not reported:** The paper doesn't disclose whether any included studies had industry funding, which could bias results.

**Limited generalizability to outdoor running:** Most RE testing was done on treadmills. Outdoor running involves wind resistance, terrain changes, and pacing variability that may affect RE differently.

For someone running their own n=1 experiment:

### What to test

Compare **HIIT** (e.g., 4 × 4 minutes at 90–95% max heart rate, with 3 minutes active recovery between intervals) versus **MICT** (e.g., 40 minutes continuous running at 70–75% max heart rate). Run each protocol for 6–8 weeks, then switch