Predicting Microbial Growth Dynamics in Commercial Cocoa-Flavored Plant-Based Milk Alternatives

This study evaluated the growth of Bacillus cereus, Escherichia coli, and Salmonella Typhimurium in commercially available cocoa-flavored plant-based milk alternatives (PBMAs) under different temperatures and examined the influence of product composition on bacterial development. Bacterial populations were monitored over 12 h at 25°C for all strains and additionally at their optimal growth temperatures (30°C for B. cereus, 37°C for E. coli, and S. Typhimurium). The nutritional composition of the PBMAs was analyzed, and principal component analysis (PCA) was performed to identify key compositional factors. Results showed that the increase in E. coli and Salmonella Typhimurium population was significantly higher at 37°C than at 25°C on all cocoa-flavored PBMAs (p < 0.05), whereas for B. cereus, the differences between 25°C and 30°C were relatively small, with statistical significance observed only for Substrate A (p < 0.05). Growth was significantly influenced by bacterial species (MS = 10.49, p < 0.001), substrate composition (MS = 0.48, p < 0.001), and interaction of both these factors (MS = 0.29, p = 0.001). The formulated regression models based on environmental factor (temperature), PCA scores (reflecting substrate composition, particularly protein, fiber, fat, sodium, and energy balance, and for B. cereus also sodium, fiber, and pH-related properties), and time, accurately reproduced bacterial growth (R² = 0.857-0.898), highlighting the combined effects of these factors. By mapping new PBMA formulations onto the tested compositional space, it may also support contamination risk assessment and contribute to ensuring consumer safety. PRACTICAL APPLICATIONS: The findings highlight that cocoa-flavored plant-based milk alternatives can support the growth of microorganisms depending on their composition and storage temperature. Understanding how nutrients such as protein, fiber, fat, and sodium drive microbial proliferation enables manufacturers to optimize formulations for improved microbial safety. The regression models developed can support predictive assessments during product design. These insights can guide industry and regulators in establishing safer handling, storage, and formulation strategies for PBMAs.