Reviewed: Pengelly A, Snow J, Mills SY, Scholey A, Wesnes K, Butler LR. Short-term study on the effects of rosemary on cognitive function in an elderly population. J Med Food. 2012;15(1):10-17.

Cognitive decline is a major threat to quality of life. Preliminary studies and traditional use have suggested that the spice rosemary (Rosmarinus officinalis, Lamiaceae) may have beneficial effects for cognitive function in older adults. Rosemary contains essential oil composed of 1,8-cineole, alpha-pinene, camphor, borneol, and carvacrol. Other constituents include phenolic diterpenes (such as carnosol and carnosic acid), flavones, the caffeic acid derivative rosmarinic acid, and the triterpene ursolic acid. It has a GRAS (generally regarded as safe) status in the United States. This randomized, placebo-controlled, double-blinded, repeated-measures, crossover study investigated the acute effects of rosemary on cognitive performance in healthy, older adults.

Subjects included 28 non-smoking adults (20 women and 8 men aged 65-90 years; mean age = 75 years) who randomly received one of four different intake levels of rosemary (750 mg, 1500 mg, 3000 mg, and 6000 mg) or placebo. While the doses for this short study were higher than normally would be consumed in the diet, the crude powder was presumed to have a similar pharmacokinetic profile to that of ordinary culinary consumption. There was a 7-day washout period between treatments. The rosemary consisted of powdered rosemary leaf (authenticated by macro- and microscopic features; McCormick and Company; Hunt Valley, Maryland) added to 458 ml of tomato juice (low sodium; Campbell’s; Camden, New Jersey). The placebo consisted of plain tomato juice.

To confound the distinction between the treatment and placebo, inactive capsules were also administered to the subjects, who were told it could be part of the treatment.

Following the one-time intake of the treatment, a battery of tests was performed at 1, 2.5, 4, and 6 hours post-consumption using the Cognitive Drug Research (CDR) computerized assessment system. Assessment included power or speed of attention; accuracy or continuity of attention; quality of working memory; quality of episodic or storage memory; and speed of memory. Mood was also assessed using the Bond-Lader visual analog scales.

There was a significant improvement in speed of memory at the lowest intake level (P=0.01) but a significant impairment at the highest intake (P<0.006) when compared with placebo. Continuity of attention and quality of working memory were also adversely affected at most intake levels. There were no effects for power of attention and quality of episodic secondary memory. Self-ratings of mood and alertness improved compared to placebo for an intake of 750 mg (P=0.01) but decreased for the 6000 mg level (P=0.02). There were no significant effects on self-rating of calmness and contentment. The mixed analysis of covariance (ANCOVA) showed there was no correlation between treatment and time for any of these findings. There were no serious adverse effects recorded for treatment or placebo.

The data show a dose-dependent effect on speed of memory; this suggests that additional work should be done at the 750 mg intake level, which is closer to normal levels of dietary intake, and also at lower doses. The authors also noted the unexpected activity of the placebo as a limitation of the study, as well as the unusually high baseline cognitive test scores of the subjects. In addition, the inability to blind the study adequately and its short-term nature were also limitations.

—Risa Schulman, PhD