Skip to main content
. 2017 Apr 26;9(5):429. doi: 10.3390/nu9050429

Table 1.

Summary of studies reporting the effects of Mg on subjective anxiety/stress in mild to moderately anxious individuals.

Author Study Design Condition Sample (N) Sex Age (year) Treatment (s) Control Duration Outcome Measure Results Effect Summary
Bourgeois et al. [74] RCT Mild anxiety (Hamilton Anxiety Scale score 10–30) N = 81 (n = 27 per condition) 20M:61F 18–65 (i) Mg 300 mg as lactate + vit B6 750 mg; (ii) Lorazepam 3 mg; (iii) (i) + (ii) combined Lorazepam 3 mg (positive verum) 6 weeks Hamilton Anxiety Scale Reduced anxiety scores in all treatments. No significant differences between treatments. x *
Scharbach [75] RCT Mild anxiety (Hamilton Anxiety Scale score 15–30) N = 133 (Treatments (i) n = 44; (ii) n = 46; (ii) n = 43) 32M:109F 18–65 (i) Mg 300 mg as lactate + vit B6 750 mg; (ii) Lorazepam 2 mg; (iii) (i) + (ii) combined Lorazepam 2 mg (positive verum) 6 weeks Hamilton Anxiety Scale Reduced anxiety scores in all treatments. No significant differences between treatments. x *
Caillard [60] RCT Mild anxiety/general anxiety disorder (Hamilton Anxiety Scale score 15–30 & general anxiety disorder (DSM III criteria)) N = 93 25M:68F x = 41 (SD = 12; 18–65) Mg 192 mg as lactate + vit B6 20 mg Placebo 6 weeks Hamilton Anxiety Scale Significant change from baseline (Total score) between groups at Day 21 (Mg + vit B6: x = 12.1 (SD = 6.0); placebo: x = 15.5 (SD = 5.8)) vs. Day 0 (Mg + vit B6: x = 21.0 (SD = 4.5); placebo: x = 22.6 (SD = 4.4); p < .03). No significant differences between Day 0 & Day 42. +
Rouillon et al., [59] RCT Mild anxiety/general anxiety disorder Hamilton Anxiety Scale score 15–30 & general anxiety disorder (DSM III-R criteria)) N = 99 (Mg n = 51; Buspirone n = 48) 38M:61F x = 37.7 (SD = 10.7; 19–65) Mg 192 mg as lactate + vit B6 20 mg Buspirone 40 mg (positive verum) 6 weeks Hamilton Anxiety Scale Decrease in anxiety scores in both treatment groups across intake. No significant difference between the efficacy of Mg + vit B6 & Buspirone. x *
Caillard [61] RCT Symptoms of functional impairment associated with anxiety or a somatic disorder (Hamilton Anxiety Scale 1; Raskin depression scale < 7; COVI anxiety scale = 7) N = 103 26M:77F x = 37 (18–65) Mg 192 mg as lactate + vit B6 20 mg Placebo 6 weeks Hamilton Anxiety Scale (somatic score) Significantly lower somatic anxiety rating after treatment at Day 21 (x = 8.4 (SD = 3.8); p = 0.004) & Day 42 (x = 6.5 (SD = 3.0); p = 0.02) vs. placebo (Day 21: x = 9.9 (SD = 2.9); Day 42: x = 7.8 (SD = 3.6)). +
Hanus et al. [72] RCT Mild anxiety/general anxiety order (Hamilton Anxiety Scale score 16–28 & somatic score ≥ 50% total score; & general anxiety disorder) DSM-III-R)) N = 264 (Treatment n = 130; Placebo n = 134) 26M:213F Placebo: x = 44.5 (18–82); Treatment: x = 44.8 (19–81) Hawthorn extract 75 mg, California poppy 20 mg + elemental Mg 75 mg (Sympathyl®) Placebo 12 weeks Hamilton Anxiety Scale Self-reported anxiety (100 mm VAS) Physician global impression Total anxiety score: Significant decrease in both conditions. Effect larger in treatment group. Mean change from baseline between Day 0 & Day 90 significantly greater in treatment group (x = −10.6 (SD = 1.2)) vs. placebo (x = −8.9 (SD = 1.2); p = 0.005). Somatic score change from baseline: Treatment (x = −6.5 (SD = 0.7)) Placebo (x = −5.7 (SD = 0.7); p = 0.054). Self-rated anxiety VAS: Mean change from baseline between Day 0 & Day 90 significantly greater in treatment group (x = −38.5) vs. placebo (x = −29.2; p = 0.005). Physician global impression: benefit > risk rating significantly greater in treatment (90%) vs. placebo (80%; p = 0.0018). +
Cazaubiel & Desor [58] RCT Mild anxiety (Hospital Depression & Anxiety Scale (HADS) score 4–12) N = 80 (Treatment n = 40; Placebo n = 40) 26M:54F Not reported Fermented cow’s milk drink (100 mL) containing milk protein hydrolysate 222 mg + Mg 48 mg (Mg form unknown) + blackberry puree Placebo 4 weeks HADS Symptom Checklist Cohen Perceived Stress Scale Vitaliano Coping scale No significant difference between treatment & placebo on study outcome measures. Post hoc analysis on restricted data (HADS anxiety subscale score 4–8, excluding scores ≥ 9) revealed significant decrease of 31% in treatment group (n = 15) vs. placebo (n = 16) on the anxiety sub-scale of the HADS (p < 0.05). + 2
Gendle et al. [76] RCT Subjective anxiety (Westside Test Anxiety Scale; normal anxiety; elevated normal anxiety; high anxiety; very high anxiety) N = 122 31M:91F x = 19.3 (SD = 1.17; 18–22) Mg 300 mg as Mg citrate Placebo (gelatin) 5 days Spielberger State-Trait Anxiety Inventory No significant difference between treatment and placebo on pre-exam anxiety rating. x

1 Total Score > 20, with sum of 2 first items < 5 & score for item 6 (depressed mood) < 2; 2 Post hoc analyses; * No difference between treatments; Mg—Magnesium; mg—milligrams; VAS—visual analogue scale; + positive treatment effect; x—no treatment effect; RCT—randomised controlled trial; Hospital Anxiety & Depression Scale—HADS; SD—standard deviation.