On the Need for Antibiotics to Reduce Subject Losses and Biases in Experiments with Aquatic Molluscs

Abstract

Biological research is frequently hampered, prevented, or biased by subject losses (Grafen, 1988; Weis, 2018). Subject losses force researchers to either repeat experiments or conduct analyses on a subset (the survivors) of the original experimental organisms. Repeating experiments has obvious financial and logistic consequences and offers no guarantee that losses will be fewer in subsequent efforts. Furthermore, data taken only on survivors is biased when mortality is correlated with variables of interest – the “missing fraction” problem (Grafen, 1988; Bennington & McGraw, 1995; Nakagawa & Freckleton, 2008). Subject losses reduce experimental sample size and balance and thereby also reduce statistical confidence, inferential power, and ultimately, the value of the research. Losses in aquatic research can stem from a wide range of causes such as inappropriate water chemistry, temperature, sudden change in physical parameters, or pathogens (Mori & Smith, 2019). Pathogens, in particular, can be a major cause of experimental subject mortality, even mass die-offs, due largely to the enclosed systems in which experiments are conducted (Kent et al., 2009; Mori et al., 2019). Yet even where pathogens cause little mortality, morbidity effects can similarly, though more cryptically, bias results (Kent et al. 2009). The best approach to protecting experiments where losses have been known to occur will often be to proactively prevent subject deaths and morbidity. Improved husbandry, including quarantines and preventive medication, can potentially reduce or eliminate subject losses and pathogen associated biases (McEwen & Fedorka-Cray, 2002). Preventive measures can introduce side-effects or bias of their own, such as gut or skin flora disruption which can impact experimental endpoints such as body weight (Carlson et al., 2017). Thus antibiotics should be considered only if benefits are likely to outweigh drawbacks for projects that are costly MALACOLOGIA, 2022, 64(2): 303–307