The Case Against Excessive Cardio for Lifters: What the Interference Effect Actually Costs

Walk into any commercial gym and you will hear some version of the same warning: cardio eats muscle. Skip the treadmill, the story goes, or you will spend your winter bulk running off the gains you paid for. The claim has a real research paper behind it, which is why it has survived four and a half decades of retelling. The problem is that almost nobody who repeats it has read the paper, and the paper does not say what the gym says it says.

The honest version is more useful and more boring. Concurrent training carries a cost. The cost is small at normal doses, concentrated in explosive strength rather than muscle size, and largely eliminated by three or four scheduling decisions that take no extra effort. Meanwhile the thing you would give up to avoid that cost, cardiorespiratory fitness, is one of the strongest predictors of how long you live that medicine has ever measured. That trade is not close.

This article makes the case against excessive cardio, with heavy emphasis on the word excessive. If you want the fat-loss comparison between interval and steady-state work, that lives in our HIIT vs. LISS breakdown. This one is about the mechanism, the actual size of the interference, and the programming rules that let you keep both qualities.

What Hickson Actually Found in 1980

Robert Hickson was an exercise physiologist at the University of Illinois at Chicago who also lifted. He noticed that his running was interfering with his strength work and, in the finest tradition of self-experimentation turned into science, he designed a study to test it. The result, published in the European Journal of Applied Physiology and Occupational Physiology in 1980, is the origin point for every "cardio kills gains" claim you have ever read.

The design was three groups over ten weeks. A strength group trained 30 to 40 minutes a day, five days a week. An endurance group trained 40 minutes a day, six days a week. The third group did both, in full, with no reduction to either program.

The findings split cleanly. VO2max climbed roughly 25% measured on a bicycle and 20% on a treadmill, and it climbed just as much in the combined group as in the endurance-only group. Endurance adaptation was untouched. Leg strength was a different story. The strength-only group added strength at a consistent rate across the full ten weeks. The endurance-only group gained essentially nothing. The combined group tracked the strength group for the first seven weeks or so, then flattened out and lost ground over the final weeks of the study.

Two details matter enormously and get dropped every time this study is retold. First, the interference was one-directional. Lifting did not compromise aerobic gains at all. Second, look at the endurance dose. Six days a week, including interval sessions at or near VO2max, layered on top of five days of heavy leg training, in subjects who were not conditioned for that workload, with no deload and no periodization. That is not "adding some cardio." That is two full-time training programs run in parallel until something broke.

AMPK vs. mTOR: The Mechanism Everyone Quotes

Once Hickson established the phenomenon, physiologists went looking for a molecular explanation, and they found a satisfying one. It goes like this.

Resistance training's main growth signal runs through mTOR, specifically the mTORC1 complex, which switches on the machinery of muscle protein synthesis. Endurance training's main adaptive signal runs through AMPK, the cell's energy sensor. When a cell burns through ATP and the AMP to ATP ratio climbs, AMPK gets phosphorylated and starts pushing the cell toward mitochondrial biogenesis and fat oxidation, which is exactly what you want from aerobic training.

The awkward part is what else AMPK does. Activated AMPK phosphorylates the tuberous sclerosis complex, TSC1/2, which sits upstream of mTORC1 as a brake. Pull that brake and mTORC1 activity falls. Written out as a diagram, endurance exercise directly inhibits the growth pathway. It is elegant, it is teachable, and it appears in every textbook chapter on concurrent training. Ken Baar's 2014 review in Sports Medicine laid out the framework and the practical implications more carefully than most.

Then human studies started testing it, and the picture got messier in a direction that favors lifters.

Apró and colleagues, publishing in the American Journal of Physiology: Endocrinology and Metabolism in 2013, had trained subjects perform a resistance session and then cycle for 30 minutes at 70% of VO2max just fifteen minutes afterward. If the AMPK story were the dominant force, that protocol should have flattened the anabolic response. It did not. mTORC1 signaling following the resistance exercise was not impaired by the subsequent endurance bout.

Work out of John Hawley's lab found the same conditional pattern from the other direction. mTOR activity was inhibited after ten maximal six-second sprints, and not after 30 minutes of moderate cycling. The molecular interference exists, but it turns on at high intensities and in energy-depleted states, and it stays quiet during the kind of moderate aerobic work most lifters actually do. Fed state, adequate carbohydrate, and moderate intensity all soften or erase it.

What Four Decades of Meta-Analyses Say

The literature since Hickson runs to hundreds of trials, which means we can stop arguing from single studies and look at pooled estimates. Four matter most.

Wilson 2012: the effect sizes and the moderators

Wilson and colleagues published a meta-analysis in the Journal of Strength and Conditioning Research covering 21 studies and 422 effect sizes. It remains the most-cited quantification of interference, and it is the source of most of the practical rules in this article.

Mean effect sizes from Wilson et al. 2012. Higher is better. Concurrent training beat endurance-only on every outcome and trailed resistance-only on every outcome.

Power took the worst of it, which makes sense. Rate of force development is the most fragile quality in the training toolbox and the first thing residual fatigue degrades. If you are a powerlifter, weightlifter, or field-sport athlete whose living depends on explosive output, you have more reason to be careful than a hypertrophy-focused lifter does.

The moderator analysis was the more actionable half of the paper. Resistance training performed alongside running produced significant decrements in hypertrophy and strength. Performed alongside cycling, it did not. And the correlations with dose were negative and clear: frequency of endurance training correlated at -0.26 to -0.35 with resistance outcomes, and duration correlated at -0.29 to -0.75. Duration was roughly twice as predictive as frequency.

Murach and Bagley 2016: the pushback

Murach and Bagley published a current-opinion piece in Sports Medicine arguing that the interference case for hypertrophy specifically was much weaker than the field assumed. Their argument was that acute molecular data had been over-extrapolated, that chronic human trials did not consistently show reduced muscle growth in concurrent groups, and that under some conditions concurrent work appeared to augment resistance-induced hypertrophy rather than blunt it.

Schumann 2022: the current best estimate

The most complete pooled analysis to date came from Schumann and colleagues in Sports Medicine, covering 43 studies of concurrent training versus strength training alone. For muscle mass, the standardized mean difference was -0.01. That is zero with extra steps. Adding aerobic training to a lifting program did not measurably compromise muscle growth.

Maximal strength showed a small negative effect. Explosive strength showed a clearer one, and the explosive-strength penalty was concentrated in studies where the aerobic work and the lifting happened in the same session. Separate the sessions by several hours and even that impairment largely dissolved.

Petré 2021 and Huiberts 2023: who it happens to

Petré and colleagues examined maximal dynamic strength development across untrained, moderately trained, and trained individuals in Sports Medicine, and found the interference was neither universal nor uniform. Huiberts, Wüst, and van der Zwaard extended this in a 2023 Sports Medicine meta-analysis looking at sex and training status. Their findings were specific: lower-body maximal strength adaptation was blunted in males but not in females, a statistically significant sex difference. Lower-body strength was blunted in untrained participants and in trained endurance athletes, and the effect did not reach significance in trained strength athletes. VO2max improvements were impaired in untrained participants but not in trained ones.

Put those together and the picture is that interference is a lower-body, male-skewed, strength-and-power phenomenon that shows up most reliably in people whose training background is endurance-heavy or nonexistent. If you are an experienced lifter adding a few conditioning sessions, you are in the population where the pooled evidence is weakest.

The Four Dials That Set the Size of the Problem

Interference is not a fixed tax you either pay or avoid. It is a dose-dependent output of four adjustable inputs. Getting all four wrong is what produced Hickson's result. Getting them right is what makes the cost disappear.

Robineau's 2016 trial deserves a closer look because it is the cleanest test of the proximity dial. Fifty-eight amateur rugby players trained for seven weeks, assigned to control, strength-only, or one of three concurrent conditions separated by 0, 6, or 24 hours, with strength always performed first. The conclusion was direct: training the two qualities with less than six hours between them was suboptimal for both neuromuscular and aerobic improvement. The authors advised coaches to avoid scheduling contradictory qualities inside a six-hour window.

Eddens and colleagues answered the order question in a 2018 Sports Medicine meta-analysis. Resistance-then-endurance beat the reverse by 6.91% for lower-body dynamic strength change. Sequence had no effect on hypertrophy, static strength, VO2max, or body fat percentage. So the order rule is narrow but free: if the sessions must touch, lift first.

The Underrated Mechanism: Fatigue, Not Signaling

Here is where most articles on this topic stop, and where the practically important explanation actually begins. The AMPK story is the one everyone teaches, and it is probably the smaller of the two mechanisms in real training.

Doma, Deakin, and Bentley published a 2017 review in Sports Medicine examining how residual fatigue from a single resistance session degrades a subsequent endurance session, and the mechanisms they identified run in both directions. They listed impaired neural recruitment patterns, reduced movement efficiency from altered kinematics and therefore higher energy cost, increased muscle soreness, and reduced muscle glycogen. Those carry-over effects last for hours to days when recovery is inadequate.

Now flip it. A 45-minute run on Tuesday leaves you with sore, damaged, glycogen-depleted legs on Wednesday. Wednesday's squat session is not blunted by a phosphorylation cascade. It is blunted because you put 315 on the bar instead of 335, or you got seven reps instead of nine, or the last two sets felt bad enough that you cut them. Do that for twelve weeks and the difference in accumulated training stimulus is enormous, and none of it required a single molecule of AMPK.

This reframing matters because it changes what you do about the problem. If interference were purely molecular, your options would be limited to timing and nutrition. Because it is largely fatigue-mediated, your options are the entire recovery toolkit: total weekly workload, calorie intake, sleep, session ordering, and mode selection. Those are levers you already know how to pull. Our guides on training volume and deload weeks cover the same math from the lifting side.

It also explains the mode finding better than the molecular story does. Running interferes more than cycling in Wilson's pooled data, and running is not more metabolically demanding per minute than hard cycling. What it is, is more eccentric. Every foot strike is a braking action, and eccentric loading is the primary driver of exercise-induced muscle damage. Cycling is concentric-dominant with almost no impact. If muscle damage and residual fatigue are the operative mechanism, that difference is exactly what you would predict.

Where "Excessive" Actually Starts

The research gives us dose-response direction and rough thresholds rather than a bright line. Combining Wilson's correlations, Hickson's protocol, and the null result in Schumann's pooled hypertrophy data, here is a defensible working model.

Notice what has to be true before you get into trouble: high frequency, long duration, impact-heavy mode, and proximity to lifting, all at once. Most lifters who believe cardio is killing their gains are doing two 25-minute treadmill sessions a week and blaming those for a plateau that is actually caused by eight hours of sleep a night becoming six, or by running the same 3x10 with the same weight for four months. Before you cut conditioning, check whether your program has progressed at all.

The Case For Cardio Is Stronger Than the Case Against

Any honest article about limiting cardio has an obligation to state what you would be giving up, because the health data here is not a close call.

Mandsager and colleagues published a retrospective cohort in JAMA Network Open in 2018 covering 122,007 patients who underwent exercise treadmill testing at the Cleveland Clinic between 1991 and 2014, with a median follow-up of 8.4 years. Cardiorespiratory fitness was inversely associated with all-cause mortality, and critically, there was no observed upper limit of benefit. Elite performers, defined as two or more standard deviations above the age and sex mean, had roughly 80% lower risk-adjusted mortality than the lowest-performing group. The adjusted mortality risk associated with poor fitness was comparable to or greater than that of coronary artery disease, type 2 diabetes, and smoking.

Read that last sentence again. Being unfit carried a risk signal on the order of smoking. Resistance training has its own excellent mortality data, and it does not substitute for aerobic capacity. They are different adaptations in different tissues.

There is a performance argument too, and it is more relevant to lifting than most people expect. Better aerobic fitness improves phosphocreatine resynthesis and clearance between sets, which means shorter effective rest periods for the same performance, which means you can complete higher training volume in the same session at the same effort. If you have ever watched someone gas out on the fifth set of squats for reasons that had nothing to do with their legs, you have watched a conditioning limitation masquerade as a strength limitation.

The general population targets from the 2018 Physical Activity Guidelines for Americans are 150 to 300 minutes of moderate-intensity aerobic activity per week, or 75 to 150 minutes of vigorous, plus muscle-strengthening activity on two or more days. Notice that the bottom of the moderate range, 150 minutes, is three 50-minute sessions or five 30-minute sessions. That is comfortably inside the zone where the concurrent training literature shows no hypertrophy cost. The health minimum and the interference threshold are not in conflict. Our article on slow steady-state cardio makes the long-term case in more depth.

How to Program Cardio Without Paying For It

Seven rules, in rough order of how much they matter.

1. Separate the sessions

Different days is best. If you must do both in one day, put at least six hours between them, which is what Robineau's data supports. Morning cardio and evening lifting, or the reverse, both work. Same-session concurrent training is where the explosive-strength penalty in Schumann's pooled data actually lives.

2. Lift first when they share a session

Eddens's meta-analysis gives lifting-first a 6.91% edge in lower-body dynamic strength change with no downside on any other outcome. The mechanism is simple: heavy compound work needs an unfatigued nervous system and full glycogen, and moderate aerobic work does not.

3. Prefer low-impact, concentric-dominant modes

Cycling, rowing, elliptical, incline treadmill walking, swimming, and sled pushes all deliver the aerobic stimulus with a fraction of the eccentric damage. Wilson's modality finding is one of the largest and most consistent moderators in the entire concurrent training literature. If you love running and it is the reason you train, keep running and accept the tradeoff knowingly. If cardio is a health box you want checked, get on a bike.

4. Cap the duration before you cap the frequency

Duration was the stronger negative correlate in Wilson's analysis. Four 25-minute sessions will cost you less than two 60-minute sessions, even though the second option has lower frequency. Short and frequent beats long and rare when muscle is the priority.

5. Put hard conditioning near lower-body training, not before it

If you run intervals or do a hard bike session, place it on the same day as your lower-body lifting session, several hours after, or the day after. Placing it the day before leg day is the worst possible slot. This concentrates fatigue rather than spreading it, leaving genuinely fresh days elsewhere in the week.

6. Feed it

Muscle glycogen is a shared resource, and the molecular interference that does exist is strongest in energy-depleted, fasted states. Carbohydrate around training blunts both the AMPK response and the recovery cost. Fasted morning cardio during a hard training block is one of the few genuinely counterproductive habits in this space. See our carbohydrate guide for the details.

7. Cut cardio last, not first

When progress stalls, the audit order is sleep, calories, program progression, lifting volume, then conditioning. Cardio is usually the smallest contributor and the first thing people amputate, partly because it is the least fun to keep.

Weekly Templates by Goal

Three worked examples. All assume four lifting days, which is the most common structure among intermediate lifters.

The fat-loss column is the only one that approaches the frequency where interference becomes a live concern, and even there the modes are low-impact and the sessions never touch heavy lower-body work in the same block. Most of the fat-loss volume is walking, which does not register as interference at all and which our walking article makes the case for. The point of that structure is to create a deficit with the least recovery cost per calorie, which is the same logic our cutting guide applies to food.

Cuts, Bulks, Sports, and Other Special Cases

During a bulk

Keep two or three easy sessions. The goal is preserving aerobic base and appetite regulation, and there is no evidence that this dose costs you anything at maintenance-plus calories. Cutting cardio to zero during a bulk is a folk tradition, not a research finding, and it usually means starting your next cut from a worse aerobic baseline and a worse work capacity.

During a cut

This is where conditioning earns its place, and where the risk of overdoing it is highest. In a deficit you have less glycogen, less recovery capacity, and a stronger temptation to add cardio instead of removing food. The interference literature was almost entirely conducted at maintenance or above, so extrapolating "cardio is basically free" into an aggressive deficit is a stretch. Add cardio gradually, protect lifting intensity as the top priority, and keep protein at the top of the range.

Strength and power athletes

Power took the largest proportional hit in Wilson's pooled data and the explosive-strength impairment in Schumann's analysis was the most robust finding of the whole meta-analysis. If your sport is measured in bar speed or vertical displacement, keep conditioning low-impact, keep it away from your key sessions, and reduce it further in the weeks before competition. General health cardio still belongs in the off-season.

Endurance athletes who lift

The interference runs one way. Hickson's endurance group gained full VO2max improvements despite the added lifting, and Huiberts's meta found VO2max impairment only in untrained participants. If you are a runner or cyclist adding strength work, the strength work is very unlikely to cost you aerobic performance and quite likely to improve your economy and injury resistance.

Beginners

Untrained participants showed the clearest interference in both Petré's and Huiberts's analyses, which sounds alarming until you remember that beginners improve at everything simultaneously and from a low base. The practical read is that a true novice should establish a lifting habit first and layer conditioning in over the following months rather than starting two demanding programs on the same Monday.

People on GLP-1 medications or in large deficits

Anyone losing weight rapidly is already fighting to hold lean mass. Adding high-volume cardio to a large deficit compounds the problem. Prioritize lifting and protein, keep conditioning modest and low-impact, and treat step count as your primary aerobic tool. Our article on GLP-1 receptor agonists covers the muscle-preservation side.

Common Mistakes

• Blaming cardio for a stalled program. Interference is real and it is not the leading cause of plateaus. Sleep, calories, and a program you have not progressed in four months all outrank it.
• Doing cardio immediately before squats or deadlifts. The single most costly scheduling choice available, and the one most commonly made because the treadmill is by the door.
• Choosing running by default. The modality effect is one of the largest moderators in the literature, and swapping to a bike or rower costs you nothing in aerobic benefit.
• Adding hours of cardio instead of adjusting food during a cut. Calories are cheaper than recovery. Use conditioning to support the deficit, not to create all of it.
• Fasted long cardio during a growth phase. Energy-depleted states are exactly where the AMPK-mediated interference is strongest and where the recovery cost is highest.
• Eliminating cardio entirely. Trading an 80% mortality risk difference for a small, controllable, mostly power-specific training cost is a bad deal by any reasonable accounting.
• Interval work every session. High-intensity intervals carry a recovery cost closer to a lifting session than to an easy bike ride. One or two hard conditioning sessions a week is plenty; make the rest easy.

The Bottom Line

The interference effect started as a genuine observation in a study that deliberately stacked two maximal programs on top of each other and watched one of them break. Four decades later, the pooled evidence says the phenomenon is real, small at ordinary doses, concentrated in explosive strength rather than muscle size, and controlled almost entirely by four variables you set yourself.

Muscle growth, on current best evidence, is essentially unaffected. Schumann's standardized mean difference of -0.01 across 43 studies is as close to a null result as this literature produces. Maximal strength takes a small hit that appears most reliably in untrained lifters, in males, and in the lower body. Power takes the real damage, and most of that penalty is a same-session effect that vanishes when you separate the work.

So the case against excessive cardio is a case against a specific combination: high frequency, long duration, high-impact mode, and stacked close to your lifting. Change any one of those and the cost drops. Change all four and it disappears into measurement noise, while you keep an adaptation that predicts mortality better than most things a cardiologist can measure.

Ride a bike for half an hour, three times a week, on days you are not squatting. That is the whole intervention. Then go back to worrying about the things that actually determine whether you grow, which remain, as always, progressive overload, adequate protein, enough calories, enough sleep, and enough years of showing up.

References

1. Hickson, R.C. (1980). Interference of strength development by simultaneously training for strength and endurance. European Journal of Applied Physiology and Occupational Physiology, 45(2-3), 255-263.
2. Wilson, J.M., Marin, P.J., Rhea, M.R., Wilson, S.M.C., Loenneke, J.P., & Anderson, J.C. (2012). Concurrent training: a meta-analysis examining interference of aerobic and resistance exercises. Journal of Strength and Conditioning Research, 26(8), 2293-2307.
3. Schumann, M., Feuerbacher, J.F., Sünkeler, M., Freitag, N., Rønnestad, B.R., Doma, K., & Lundberg, T.R. (2022). Compatibility of concurrent aerobic and strength training for skeletal muscle size and function: an updated systematic review and meta-analysis. Sports Medicine, 52(3), 601-612.
4. Murach, K.A., & Bagley, J.R. (2016). Skeletal muscle hypertrophy with concurrent exercise training: contrary evidence for an interference effect. Sports Medicine, 46(8), 1029-1039.
5. Petré, H., Hemmingsson, E., Rosdahl, H., & Psilander, N. (2021). Development of maximal dynamic strength during concurrent resistance and endurance training in untrained, moderately trained, and trained individuals: a systematic review and meta-analysis. Sports Medicine, 51(5), 991-1010.
6. Huiberts, R.O., Wüst, R.C.I., & van der Zwaard, S. (2024). Concurrent strength and endurance training: a systematic review and meta-analysis on the impact of sex and training status. Sports Medicine, 54(2), 485-503.
7. Eddens, L., van Someren, K., & Howatson, G. (2018). The role of intra-session exercise sequence in the interference effect: a systematic review with meta-analysis. Sports Medicine, 48(1), 177-188.
8. Robineau, J., Babault, N., Piscione, J., Lacome, M., & Bigard, A.X. (2016). Specific training effects of concurrent aerobic and strength exercises depend on recovery duration. Journal of Strength and Conditioning Research, 30(3), 672-683.
9. Doma, K., Deakin, G.B., & Bentley, D.J. (2017). Implications of impaired endurance performance following single bouts of resistance training: an alternate concurrent training perspective. Sports Medicine, 47(11), 2187-2200.
10. Apró, W., Wang, L., Pontén, M., Blomstrand, E., & Sahlin, K. (2013). Resistance exercise induced mTORC1 signaling is not impaired by subsequent endurance exercise in human skeletal muscle. American Journal of Physiology: Endocrinology and Metabolism, 305(1), E22-E32.
11. Baar, K. (2014). Using molecular biology to maximize concurrent training. Sports Medicine, 44(Suppl 2), S117-S125.
12. Coffey, V.G., & Hawley, J.A. (2017). Concurrent exercise training: do opposites distract? Journal of Physiology, 595(9), 2883-2896.
13. Gergley, J.C. (2009). Comparison of two lower-body modes of endurance training on lower-body strength development while concurrently training. Journal of Strength and Conditioning Research, 23(3), 979-987.
14. Mandsager, K., Harb, S., Cremer, P., Phelan, D., Nissen, S.E., & Jaber, W. (2018). Association of cardiorespiratory fitness with long-term mortality among adults undergoing exercise treadmill testing. JAMA Network Open, 1(6), e183605.
15. U.S. Department of Health and Human Services. (2018). Physical Activity Guidelines for Americans, 2nd edition. Washington, DC.