November 21, 2014

My New Study on Fasted Cardio and Fat Loss: Take Home Points

In the late 1990’s, Bill Phillips authored “Body for Life,” which went on to become one of the biggest selling fitness books of all time. In the book, Phillips claimed that performing 20 minutes of high-intensity aerobic exercise (HIIT) after an overnight fast has a greater effect on fat loss than an hour of cardio performed following consumption of a meal. The rationale for the hypothesis was based on research showing that low glycogen levels cause your body to shift substrate utilization away from carbohydrates, thereby allowing greater mobilization of stored fat for energy.

While the theory that fasted cardio is superior for fat loss is certainly intriguing, it is based on an extrapolation of findings that might not translate into practice. Several years ago I authored a review of literature that discussed the contradictions of the research on the topic. While my review highlighted a number of inconsistencies that suggested fasted cardio might not work as claimed, one little issue continued to nag at me: The entire debate was based on acute data; no study had actually investigated the effects of fasted cardio on body fat when subjects were in an energy-deficit sufficient to produce weight loss.

Until now…

My lab recently carried out a controlled longitudinal trial designed to achieve clarity on the topic. The paper titled, Body composition changes associated with fasted versus non-fasted aerobic exercise was just published in the Journal of the International Society of Sports Nutrition. Here is a rundown of what we found along with a discussion of relevant practical implications.

What We Did
Twenty young non-obese (BMI < 30) were recruited to participate in the study. Prior to the intervention, subjects were tested for body composition (weight, body fat percent, fat mass, fat-free mass, and waist circumference) using a Bod Pod (i.e. air displacement plethysmography), then pair-matched based on initial body mass measurements and randomly assigned to 1 of 2 groups: a fasted training (FASTED) group that performed exercise after an overnight fast (n =10) or a non-fasted training (FED) group that consumed a meal prior to exercise (n =10). These meals were provided in the form of a shake (Pursuit Recovery by Dymatize Nutrition) that contained 250 calories consisting of 40 g carbohydrate, 20 g protein, and 0.5 g fat. Training was carried out 3 days a week for 1 hour per session on a treadmill. Subjects performed a 5 minute warm-up followed by 50 minutes of walking/jogging at 70% max heart rate. A 5-minute cool down was then provided to end the session. We chose this protocol because evidence shows that lipid oxidation during fasted aerobic exercise is maximized during low-to moderate-intensity steady-state cardio – at higher intensities, the fasted condition allows for an acutely greater lipolysis but the overall oxidation rate is similar to the fed condition because more free fatty acids are available than can be oxidized (in fairness to Bill Phillips, this research came out after publication of his book). Subjects were provided with customized meal plans intended to bring about a 500-calorie deficit. The meal plans were flexible so that subjects had wide choices of their preferred foods. Protein was kept high to help ensure preservation of lean body mass. Subjects recorded their food in an online web-based program ( on a daily basis so that dietary intake could be continually monitored. Ongoing nutritional counseling was provided to subjects throughout the study period to promote compliance and adherence. After 4 weeks, the subjects were retested on all body composition measures. What We Found
Both groups lost a statistically significant amount of weight (1.6 kg vs. 1.0 kg in the FASTED vs. FED groups, respectively) and fat mass (1.1 kg vs. 0.7 kg in the FASTED vs. FED groups, respectively). However, no significant differences were noted between groups in any of the body composition outcomes.

Reconciling Findings with Practical Implications
On the surface, it might seem that the fasted cardio group had a slight advantage in terms of weight loss and fat loss. I’ve seen comments on social media to this effect, claiming a hidden “trend” for a benefit to fasted cardio that our study simply was underpowered to detect.

Fact is, the claim is unsubstantiated.

The p-values (a determinant of the probability that results were due to random chance) were *highly* insignificant, averaging 0.8-0.9 for the various body comp outcomes. Moreover, differences in effect sizes (a measure of the magnitude of the effect) between groups were negligible, further indicating a lack of differences. On top of all that, the FASTED group had somewhat higher (non-significant) baseline body fat percentage, providing a potential advantage for slightly greater fat loss. There are certainly times that I have seen and reported trends in studies I’ve carried out where it was apparent that the small sample size obscured significant differences. That was most definitely NOT the case here. Based on the findings, any differences noted would be attributable to chance — I’d say with a high level of confidence that the sample size was not an issue in this regard.

So does this mean the case is closed and that fasted cardio is worthless for fat loss?

Not necessarily.

As with every study, there were a number of limitations that must be taken into account when drawing evidence-based conclusions.

For one, the study spanned only four weeks in duration. Certainly this is a sufficient period of time to realize significant reductions in weight and fat mass (as was demonstrated here), but it remains possible that very slight differences between conditions *might* take longer to manifest. We limited the duration of the study in an effort to ensure that the subjects adhered to the diet and exercise protocol (longer term trials in young college students can be problematic with respect to adherence). Our study was not funded, so we could not offer remuneration as an incentive for sticking with the program. Ideally a 16 week protocol would better determine if any small effects would ultimately become significant.

Another potential confounding issue was the use of pre-menopausal women as subjects. Monthly menstrual cycles can influence body weight due to alterations in fluid balance. The fact that pre- and post-testing was conducted exactly one month apart would seem to control for any issues in this regard. However, some women have irregular menses and we cannot rule out the possibility that such fluctuations influenced results.

I will also note that subjects lost slightly less weight than anticipated. This seems to be due to consuming more calories than prescribed in the meal plans. Despite our best efforts to counsel subjects on what to eat and providing detailed instruction on how to to record foods in the online database, the subjects apparently under-reported their nutritional intake. That said, analysis of food diaries indicates that under-reporting was equally distributed between groups and thus this should not have affected overall results. Whether a larger caloric deficit would have provided an advantage to one condition versus the other is open to debate.

Finally, our findings are specific to young, non-obese women and cannot necessarily be generalized to other populations. It has been speculated that the true benefit of fasted cardio is specific to very lean individuals, such as pre-contest bodybuilders, who are trying to lose that last pound or two of stubborn fat. We cannot rule out such a possibility. I’ll note, however, that several of our subjects were off-season track athletes who were quite lean. In fact, four of the subjects (two in each group) had body fat levels that would be considered very low for women (13-16%). When analyzing the results of these four subjects, there was no evidence whatsoever that the fasting condition conferred any benefits. Admittedly this is a tiny sample and certainly cannot be taken as proof of anything. Nevertheless, it does provide a more controlled, objective perspective into potential benefits of fasted cardio than the usual “it worked for me” claims that lack any level of control or objectivity.

Bottom Line
A single study is simply a piece in an evidentiary puzzle and can never considered the final word on a topic. What I do think is clear from our study, however, is that if there are any benefits from fasted cardio (still highly equivocal), they would be minor at best. So the best advice for those who are simply looking to get lean is to focus on total energy and macronutrient balance; whether you perform cardio fasted or fed should depend entirely on preference.

On the other hand, it remains possible that a small benefit could be seen by performing fasted cardio. If such an effect does exist, it would seem to be only meaningful to someone who is competing in a bodybuilding or physique competition, where minute differences in fat mass could make the difference between winning or losing a competition. I will point out, however, that it also is conceivable fasted cardio could have a negative effect in this regard. A recent study by Paoli et al showed that lipid utilization over 24 hours was actually higher when eating prior to cardio as opposed to remaining fasted. Thus, the best advice here would be to experiment and try to objectively determine what works best for you as an individual.


  1. […] study shows no difference in body composition between fasted and non-fasted workouts. Here are some worthwhile take home points from Brad Schoenfeld, one of the study’s authors.  (Journal of the International Society of […]

    Pingback by nutrition health fitness research update — November 21, 2014 @ 2:07 pm

  2. Fantastic work, great to see studies translated for mass consumption like this. Is there any hope of capturing rebound weight gain in these volunteers?

    Comment by Daniel commane — November 21, 2014 @ 6:28 pm

  3. Amazing as always Dr Broenfeld 😉
    It’s so good to see a researcher be able to objectively critique his own study and provide objective information to the lay person. Well done

    Comment by Jason — November 21, 2014 @ 7:47 pm

  4. Thanks for insight. Actually read from suppversity blog that you were going to conduct this experiment:)

    Comment by Amir — November 24, 2014 @ 11:39 am

  5. […] 6) My New Study on Fasted Cardio and Fat Loss: Take Home Points via Brad Schoenfeld […]

    Pingback by Good Fitness Reads of the Week: 11/30/2014 | — November 30, 2014 @ 3:56 pm

  6. Excellent information Brad…Very well done! Wouldn’t have expected anything less. You are a credit to health and fitness!

    Comment by Mark T. Cuatt — December 6, 2014 @ 11:40 am

  7. […] Brad Schoenfeld talks on his new study — My New Study on Fasted Cardio and Fat Loss: Take Home Points. […]

    Pingback by Body composition changes associated with fasted versus non-fasted aerobic exercise — January 12, 2015 @ 9:23 am

  8. […] its fat stores as energy which will thus enhance fat loss. As incredible as this sounds, a recent study was carried out over 4 weeks with one group exercising from a fasted state and the other group […]

    Pingback by Intermittent Fasting: A Helpful Way to Finally Eat Less – FNDfitness — February 8, 2015 @ 7:21 pm

  9. […] Fasted cardio & fat loss […]

    Pingback by Why Fasted Cardio Sucks - Built To Dominate — March 24, 2015 @ 2:13 pm

  10. […] consiguiron una pérdida de peso y grasa significativa. De todas formas Brad, en su propio Blog (21 Nov 2014) considera que podría ser posible que un pequeño beneficio sea consiga con el cardio en ayunas. […]


  11. […] […]

    Pingback by The big debate - breakfast BEFORE or AFTER my morning weights session? — January 10, 2016 @ 9:54 am

  12. Great post. I foung your blog and i like it!
    Thanks from Sapin.

    Comment by Entrenador personal madrid — March 6, 2016 @ 4:35 pm

  13. “A single study is simply a piece in an evidentiary puzzle and can never considered the final word on a topic.” This sentence say everything. We are all just humans, but also we can be very different, in every point, also in energy expenditure and substrat utilization in varous situations.

    Brad, you probably Know the work of Claude Bouchard, lot of interesting ideas and rearches are described in book he edited, Obesity and Physical Activity. If you dont have it, I warmly recomend some chapters from it 🙂

    Comment by iver — November 23, 2016 @ 8:08 pm

RSS feed for comments on this post.

Sorry, the comment form is closed at this time.