Effects of Intermittent Fasting Promising But Poorly Studied

Benjamin D. Horne, PhD, MPH, FAHA, FACC Director, Cardiovascular and Genetic Epidemiology, Intermountain Heart Institute Adjunct Assistant Professor, Genetic Epidemiology Division, Department of Medicine, University of UtahMedicalResearch.com Interview with:
Benjamin D. Horne, PhD, MPH, FAHA, FACC
Director, Cardiovascular and Genetic Epidemiology, Intermountain Heart Institute
Adjunct Assistant Professor, Genetic Epidemiology Division, Department of Medicine,
University of Utah

Medical Research: What is the background for this study? What are the main findings?

Dr. Horne:  A rapidly expanding set of results from animal studies exists regarding the effects of intermittent fasting among animals. Many pilot studies of intermittent fasting have been performed now, too, but the body of literature regarding the human response to fasting is incomplete. This study sought to determine how preliminary and incomplete the evidence is in humans regarding the health benefits and the harmful side-effects of intermittent fasting. The purpose was to determine how reliable the evidence is that people should be engaging in fasting to improve their health. In particular, one major concern is that diet gurus and even some scientists are marketing intermittent fasting to the public through diet books and other methods that produce additional income for them, but it is unclear whether their claims can be supported.

The main findings of the study are that only three controlled clinical trials of intermittent fasting in humans have been published that were designed rigorously and included a control group in addition to the fasting intervention arm of the study. One of the three had a pre-specified primary outcome (weight loss) and another used the Bonferroni correction to account for inflated false positive results due to multiple hypothesis tests (the third study unfortunately did neither, which is the common approach in human studies of intermittent fasting). These three controlled trials only used surrogate or intermediate endpoints, though, such as weight, cholesterol, or other risk factors for disease. The three trials also have substantial limitations, including small sample sizes (~30 people total), a fasting regimen that was studied for less than three months, and no evaluation of clinical safety outcomes. Only one of the three trials was registered on ClinicalTrials.gov, a site instituted as part of the FDA Modernization Act and a pre-requisite for trials that are published in scientifically sound medical journals and are to be reviewed by the FDA.

Two observational studies of the association of intermittent fasting with clinical events (i.e., coronary artery disease diagnosis and diabetes diagnosis) also have been published. These two studies included 200 patients and 445 patients. They were performed in patients drawn from a general population in which a large proportion of people engage in fasting intermittently over a period of decades. These observational studies provide the only evidence that fasting is associated with a lower risk of disease outcomes. Such studies are limited, though, by adjustment for only known or measured confounders, making it possible that some important factors may be unobserved in the studies that would account for the fasting benefit that was observed. No randomized controlled clinical trial of fasting for clinical events or disease outcomes has been performed, though, thus these five studies are the whole body of reliable evidence that intermittent fasting is beneficial to humans.

Medical Research: What should clinicians and patients take away from your report?

Dr. Horne:  Clinicians and the public should understand from the results of this study that intermittent fasting is a promising but poorly studied behavior. Fasting may provide some health benefit, but it does not have the rigorous and extensive data at this time that are needed to be able to recommend fasting as a health intervention. Substantial additional data are needed from many additional well-designed, rigorously controlled, clinically-relevant trials of intermittent fasting prior to utilizing it to improve health in a way that avoids unexpected and unwanted harmful side-effects.

The recent revision of dietary guidelines regarding consumption of fats and cholesterol that have been in place for more than 35 years highlights these concerns. The guidelines to avoid fats and cholesterol in the diet may have contributed substantially to the obesity epidemic we are fighting today wherein a rush to make a recommendation to the public many years ago about fats and cholesterol ultimately caused more harm than benefit to the health of members of the public. In place of using exciting animal research and a few poorly-designed human studies to change dietary practices, methodical and reasoned approaches to evaluating intermittent fasting are needed that are based on established scientific and clinical methodologies for developing reliable data from which high-quality guidelines and policies can be derived.

Further, in reference to the intermittent fasting fad diets that are being marketed today, it is unlikely that fasting twice per week or more is healthy for the average individual. While weight loss is a given if a substantial decrease in caloric intake occurs (consider the logical extreme of fasting 7 days per week: yes, you will lose weight, but at what cost?), such excessive amounts of fasting are not sustainable over the long-term. Once someone stops following such intense fasting regimens, the weight will be gained again without some compensatory change in behavior such as changing the type of food that is consumed or increasing the amount of physical activity. Further, the weight that is gained back is likely to be unhealthy tissue, not lean muscle. Further, the stress that fasting excessively places on the body may also damage tissues and organs, making them more susceptible to disease or directly causing harm to their function. Care should be taken by anyone considering the use of intermittent fasting as a health intervention and consultation with a wise clinician is recommended prior to beginning an intense fasting regimen such as fasting twice per week or more.

Medical Research: What recommendations do you have for future research as a result of this study?

Dr. Horne:  Additional, well-controlled randomized clinical trials of intermittent fasting are needed. These studies should evaluate not weight loss but changes in blood pressure, cholesterol, blood sugar, cognitive function, and other outcomes. Separate studies should be performed for each of those outcomes, with one outcome per study pre-specified as the primary outcome on which the success of the study rests. Registering those studies on ClinicalTrials.gov prior to the start of the study, as required by reputable medical journals, will help ensure the validity and reliability of the study results. In those studies, secondary hypothesis tests that suggest that fasting influences other outcomes should either be tested prospectively in additional studies or should at least have their statistical significance corrected for multiple comparisons. These new studies should be non-pilot studies that enroll hundreds or thousands of individuals and they should also evaluate safety of fasting, not just its effectiveness. Little is known about the intermediate-term and long-term harms that fasting may cause.

Furthermore, clinical events should be studied as the outcomes in intermittent fasting studies so that direct evaluation of the diseases that we intend to prevent can be performed. Clinical relevance is substantially limited without the study of clinical event outcomes.

Once definitive data are established in humans that intermittent fasting does more than cause weight loss, including possibly reducing the risk of diabetes and coronary disease, additional research will be needed. Little is known about the optimal frequency or duration for each fasting episode or how long through one’s life it should be practiced. Recommendations for practicing fasting and for integrating it into clinical guidelines and policy should be held back until definitive efficacy and safety data are available.


Benjamin D Horne, Joseph B Muhlestein, and Jeffrey L Anderson

Health effects of intermittent fasting: hormesis or harm? A systematic review Am J Clin Nutr ajcn109553; First published online July 1, 2015. doi:10.3945/ajcn.115.109553

[wysija_form id=”3″]

Benjamin D. Horne, PhD, MPH, FAHA, FACC (2015). Effects of Intermittent Fasting Promising But Poorly Studied