The Effects of a Low-Fat/High-Fiber Diet on Sex Hormone Levels and Menstrual Cycling in Premenopausal Women: A 12-Month Randomized Trial (the Diet and Hormone Study)

Peter H. Gann*, Robert T. Chatterton, Susan M. Gapstur, Kiang Liu, Daniel Garside, Sue Giovanazzi, Kim Thedford, Linda Van Horn

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

54 Scopus citations


BACKGROUND. Reduction of cumulative exposure to endogenous ovarian steroid hormones is a postulated method for reducing the risk of carcinoma of the breast and other malignancies. Although there are data from trials evaluating the effect of low-fat and high-fiber diets on sex hormone levels in premenopausal women, to the authors' knowledge none of these trials has combined a relatively large number of participants, follow-up of > 2-3 months, parallel controls receiving a usual diet, and careful timing of blood sampling within the menstrual cycle. METHODS. A total of 213 healthy women, ages 20-40 years, were randomly assigned to follow their usual diet or to adopt an isocaloric diet with goals of 20% calories as fat, total fiber of 25 g/day, and at least 8 fruit or vegetable servings per day. Serum levels of total estradiol (E2), sex hormone-binding globulin (SHBG), non-SHBG-bound estradiol (NSBE2), SHBG, and progesterone were evaluated during a menstrual cycle at baseline, and at 4 cycles (C4) and 12 cycles (C12) after the start of the intervention. Serum was collected during each test cycle 7-9 days after the detection of an luteinizing hormone peak in the urine. One hundred eighty-nine women provided serum at C4 and 176 women at C12. RESULTS. Serum E2 decreased by an average of 7.5% or 7.8 pg/mL (95% confidence interval [95% CI], -16.0-0.04) at C12 in the intervention group, versus a decrease of 0.9% or 0.9 pg/mL (95% CI, -9.5-7.7) in the control group (the P value for the difference between the treatment groups was 0.25). Results for NSBE2 were very similar to those for total estradiol. There were no material effects found to result from intervention with regard to SHBG or progesterone. The results did not differ by baseline age, body mass index, or baseline hormone level above or below the median, and were not likely to be affected by weight change, which amounted to a mean loss of only 0.23 kg in the diet group versus a gain of 0.17 kg in the control group. The decrease in serum E2 associated with intervention was not greater when subjects were stratified by self-reported adherence to the dietary goals. CONCLUSIONS. The results of the current study suggest that the effects of this isocaloric low-fat, high-fiber diet pattern on circulating ovarian steroids were modest or nonexistent. However, the observed 7.5% reduction in estradiol could have biologic significance if it persisted over many years. Moreover, under-estimation of the true dietary effect could have occurred because of incomplete adherence to assigned diets. Weight loss and weight control through midlife could be a more effective and feasible approach to dietary intervention in reducing the risk of breast carcinoma.

Original languageEnglish (US)
Pages (from-to)1870-1879
Number of pages10
Issue number9
StatePublished - Nov 1 2003


  • Breast neoplasms
  • Diet
  • Estradiol
  • Fat
  • Fiber
  • Progesterone
  • Sex steroids

ASJC Scopus subject areas

  • Oncology
  • Cancer Research


Dive into the research topics of 'The Effects of a Low-Fat/High-Fiber Diet on Sex Hormone Levels and Menstrual Cycling in Premenopausal Women: A 12-Month Randomized Trial (the Diet and Hormone Study)'. Together they form a unique fingerprint.

Cite this