Food additive mixtures and incident elevated depressive symptoms in the NutriNet-Santé cohort

Background

Current safety evaluations assess food additives individually and do not account for real-world mixture exposures or chronic disease endpoints. We assessed associations between food additive mixture exposure and incident elevated depressive symptoms.

Methods

We analysed 36,516 participants (50 ± 14 years; 77% women) from the French NutriNet-Santé cohort (2009–2022) with ≥2 24-hour dietary records, no depression at baseline, and ≥1 follow-up CES-D. Incident elevated depressive symptoms were defined using sex-specific CES-D cut-offs (≥17 men; ≥23 women). Additive intakes (mg/d) were estimated from 24-hour dietary records with brand-specific data, linked to composition databases and laboratory analyses. Mixtures were derived using nonnegative matrix factorisation, and associations assessed using multivariable Cox models (age timescale) per 1-SD increase.

Results

Over a median 6 years, 4,605 cases occurred. Five mixtures were identified. Higher exposure to mixture 5 (characterised mainly by acidity regulators, sweeteners, colours, and selected emulsifying, stabilising, texturising, glazing, and preservative agents) was associated with higher incidence of elevated depressive symptoms (HRper 1-SD 1.06, 95% CI 1.03–1.08; P < 0.001), while mixtures 1–4 were not. Exploratory analyses suggested interactions between mixture 5 constituents, and associations persisted after accounting for individual additives.

Conclusions

Higher exposure to a commonly consumed food additive mixture was associated with a higher incidence of elevated depressive symptoms, with evidence consistent with a mixture effect beyond individual additives. These findings suggest that additive mixtures may represent a modifiable risk factor, warranting replication and experimental studies to clarify causality and mechanisms, and informing future safety evaluations and early policy consideration to reduce unnecessary exposure to nonessential additives.

Key messages

• This global-first prospective study quantified population exposure to food additive mixtures and found higher exposure to one mixture was linked to higher incidence of elevated depressive symptoms.

• Findings suggest additive mixtures, not just individual additives, may be a modifiable risk factor for depressive symptoms and relevant to food safety frameworks, helping guide early policy action.