Abbasi · Journal of research in medical sciences : the official journal of Isfahan University of Medical Sciences 2012 · double-blind randomized placebo-controlled trial · n=46

The effect of magnesium supplementation on primary insomnia in elderly: A double-blind placebo-controlled clinical trial.

Cited 114 times in the scientific literature.

Level 2 - randomized trial

Individual double-blind randomized placebo-controlled trial

PubMed 23853635 · record verified 2026-08-29

What was done

A double-blind randomized controlled trial evaluated 46 elderly participants with primary insomnia allocated to receive either 500 mg magnesium daily or a placebo for 8 weeks. Assessments at baseline and after 8 weeks included the Insomnia Severity Index (ISI), sleep logs, physical activity, 3-day 24-hour dietary recalls, and blood biomarkers (serum magnesium, renin, melatonin, and cortisol).

What was found

Compared to placebo, magnesium supplementation resulted in statistically significant increases in sleep time (P = 0.002), sleep efficiency (P = 0.03), serum renin (P < 0.001), and serum melatonin (P = 0.007). It led to significant reductions in ISI score (P = 0.006), sleep onset latency (P = 0.02), and serum cortisol concentration (P = 0.008). Reductions in early morning awakening (P = 0.08) and increases in serum magnesium (P = 0.06) were marginally significant. Total sleep time did not differ significantly between groups (P = 0.37). Exact numerical values for these outcomes were not reported in the abstract.

Why it matters

This trial suggests that 500 mg daily magnesium may serve as an affordable, low-risk nutritional intervention to improve subjective sleep quality and modulate neuroendocrine markers in older adults suffering from insomnia.

Limits

The study had a small sample size (n = 46) and short duration (8 weeks). Sleep outcomes relied on subjective logs and surveys rather than objective measures like polysomnography or actigraphy. The abstract did not report absolute numerical values or effect sizes, providing only p-values.

Cited by