Compared with people who consume low levels of dietary fiber, those who consume high levels have decreased rates of all-cause, cardiovascular, infectious, and respiratory mortality, according to a study published online Feb. 14 in Archives of Internal Medicine.
Higher fiber intake also was related to decreased cancer mortality, but only among men, said Yikyung Park, Sc.D., of the nutritional epidemiology branch of the National Cancer Institute, Rockville, Md., and her associates.
The study findings suggest that choosing fiber-rich foods may have across-the-board health benefits, and may reduce the risk of premature death from all causes, they noted.
Many observational studies examining the effect of dietary fiber on mortality have been limited by small sample sizes, narrow ranges of fiber intakes, and inadequate control for confounding factors; and therefore many of the studies have produced inconsistent results.
The investigators examined the issue using data from a large cohort - participants in the National Institutes of Health/American Association of Retired Persons Diet and Health Study.
The NIH/AARP study involved more than 567,000 men and women, aged 50-71 years, residing in six states and two metropolitan areas, who completed detailed questionnaires about their diets in 1995-96 and were followed for a mean of 9 years. For this analysis, the responses of 219,123 men and 168,999 women were assessed.
There were 20,126 deaths among men and 11,330 among women during follow-up, including 7,845 deaths due to cardiovascular disease (CVD) and 13,171 cancer deaths. Dietary fiber intake ranged from 11 grams per day on the low end to 29 grams per day on the high end.
Compared with people in the lowest quintile of fiber consumption, those in the highest quintile had a 22% lower risk of death from all causes, Dr. Park and her colleagues reported (Arch. Intern. Med. 2011 Feb. 14 [doi:10.1001/archinternmed.2011.18]).
Because smoking status was a strong confounder in the analysis, the data were analyzed again after subjects were stratified by smoking status. Among people who had never smoked, the inverse association between fiber consumption and total mortality remained significant for men and women. It also remained significant among former smokers and current smokers.
Similarly, the inverse association between fiber intake and mortality remained strong across all categories of body mass index.
Men in the highest quintile of fiber consumption had a 24%-56% lower risk of death from CVD, cancer, infectious disease, or respiratory disease than did men in the lowest quintile. Women in the highest quintile of fiber consumption had a 34%-59% lower risk of death from CVD, infectious disease, or respiratory disease, but there was no inverse association between fiber intake and cancer mortality in women.
This discrepancy may be explained in part by differences between men and women in leading organ sites for cancer deaths, the investigators noted.
In a further analysis, the data were adjusted to account for the use of aspirin therapy, the presence of hypertension, and the presence of cholesterolemia. "No appreciable differences" in the link between dietary fiber and mortality were found.
When the data were analyzed by fiber source, high dietary fiber from grains was significantly associated with decreased mortality, while fiber from vegetables and beans showed a weaker association and fiber from fruits showed no association.
Several plausible mechanisms have been suggested to explain the benefits of dietary fiber. Fiber's anti-inflammatory properties may explain its protective effect across several types of disease, Dr. Park and her associates proposed.
However, it also is possible that high consumption of dietary fiber in this study was simply a marker for a healthy diet and lifestyle.
Study subjects in the highest quintile of fiber intake were more likely to have a higher education, a lower BMI, and a higher level of physical activity, and to have rated their health as being very good or excellent, compared with subjects in lower quintiles. Although these factors were controlled for in multivariate analyses, some residual confounding may have persisted, the researchers said.