7/7

This highlights the importance of training specificity (i.e. long runs) and higher mileage for sustaining economy during prolonged exercise – key for events from 10k to marathon.

🔄 Share & discuss!

📄 Free full text (AoP): journals.lww.com/acsm-msse/ab...

6/7

Take-home messages:

Even with identical performance levels, training matters for durability.

Runners regularly including long runs – and accumulating higher mileage – had:

✅ More durable running economy
✅ Smaller neuromuscular declines

5/7

When we merged data from both groups, RE durability correlated:

📉 strongly to weekly longest run (r=-0.67; p<0.001)

📉 and moderately to weekly running distance (r=-0.48; p=0.038)

But not to declines in neuromuscular function.

4/7

Neuromuscular function followed a similar trend:

🦵🏼 SDT showed larger drops in maximal squat force (-19% vs -12%)

⬆️ SDT lost CMJ performance (6.6%), while LDT maintained it

3/7

During a 90 min run at LT:

- Running economy deteriorated nearly 50% less in LDT than SDT (+3.1% vs +6.0%)

- Differences emerged after 60min and grew with time

2/7

We matched well-trained runners for 10k time (39 min), V̇O₂max (~58 ml/kg/min) & LT speed (12-13 km/h)

But differed in training characteristics:
- LDT: regularly running ≥90min
- SDT: always run <70min

LDT also ran more (51 vs 30 km/wk) & had better fresh running economy

5/5

❗️Final message

Researchers should consider, quantify, and report TL.

If you’re assessing training interventions, matching TL may help ensuring results reflect their effect, without being confounded by unintended differences in the dose of training prescribed.

4/5

We also highlight why TL matching might not be essential. For instance when:

- Studying real-world training programmes

- Investigating outcomes less sensitive to dose (like enjoyment or adherence).

3/5

Our paper explores how TL mismatches can skew results, and when matching TL is likely beneficial.

We provide 3 practical examples in areas where TL warrants consideration.

And briefly discuss how an increased TL carries differences depending on participants characteristics

2/5

Training load (TL) can be associated to the "dose" of exercise.

When groups in a study receive different doses, it becomes unclear whether the outcomes are due to the type of training, or simply more training.

An issue for interpreting effectiveness in some contexts.

8/8

‼️Results matter for durability studies assessing interventions & longitudinal athlete profiling.

🔑 Finding: Even subtle differences in RE durability between conditions are likely detectable.

🔁 Share if you find this interesting.

🆓 Open access
📄 onlinelibrary.wiley.com/doi/10.1111/...

7/8

Summary:

✅ RE (expressed as OC or EC) is reliable during prolonged running
✅ VE, RER, and HR also consistent
⚠️ BLa and RPE less so, esp. >60 min

4/5 PhD studies published. As always thanks to Rich Blagrove and Jonathan Folland.

6/8

Blood lactate (BLa) was the least reliable:

🔹 ICC = 0.48–0.94
🔹 CV = 5–16%

Reliability dropped after 60 min.

This likely reflects limits in analyser precision more than physiology.
Caution warranted when interpreting small BLa changes during prolonged exercise!

5/8

RPE was moderately reliable, with:

🔹 ICCs = 0.41–0.72
🔹 CV = 4.2–6.0%
No effect of time on CV – fatigue didn’t make it less reliable.

Important: subjectivity means RPE isn’t as stable as physiological data.

4/8

Heart rate (HR) was also highly reliable:

🔹 ICCs = 0.82–0.92
🔹 CV = ~1.1%
🔻 Slightly lower in trial 2 at later timepoints; possibly due to familiarisation.

Still well within typical day-to-day HR variability.

3/8

Ventilation (VE) and RER also showed strong reliability:

🔹 VE: ICC = 0.96–0.97, CV < 3.6%
🔹 RER: ICC = 0.78, TE ≤ 1.9%

These markers are consistent across 90 min of heavy-intensity running.

2/8

Both Energy Cost (EC) and Oxygen Cost (OC) showed excellent reliability across all timepoints:

✅ ICCs: 0.96–0.99
✅ CVs: 0.6%–1.2%
✅ TE: ≤1.4%

No decline in reliability over time.

‼️RE durability is measurable and trustworthy.

I think a possible solution could be having a testing session as an outdoor long run with pre-post assessment of running economy and speed at LT (this could be done on a separate day).

sLT is likely to capture multiple changes under a single variable, and testing it may not be too intense.

Thank you Jamie!

I think a thorough durability testing for high-level marathon is quite challenging - as you already pointed out. Athletes more trained than our cohort will likely display smaller changes, so either duration or intensity would need to increase..