Abstract
The controlled experiment is one of the most powerful tools available in biomedical research, and thousands of such experiments are conducted annually. However, the basic principles of designing such experiments do not seem to be widely understood. If the experimental material is heterogeneous, then the treated and control groups may not be identical at the start of the experiment, even if large sample sizes are used. Yet genetically heterogeneous ‘outbred’ stocks of laboratory mice and rats are widely used in the pharmaceutical industry in spite of the availability of more homogeneous inbred strains. The use of such stocks is justified on two grounds: that humans are also heterogeneous, and that such stocks offer a wide range of phenotypes for toxicological screening. Both arguments are spurious. It is not necessary for atoxicological model of humans to be a goodgenetic model. Models do not have to resemble their target in every respect. Were they to do so they would no longer be models. The genetic heterogeneity within an outbred stock simply creates ‘noise’ when comparing treated and control groups, and prevents more exact matching of groups before the experiment starts. Phenotypic heterogeneity can be achieved without this disadvantage using a factorial experimental design involving more than one strain, with groups exactly matched genetically, but without increasing total numbers.
The lack of understanding of good experimental design seems to extend to other areas, with few research workers using techniques such as blocking and factorial experimental designs which have been known for over 60 years, and which are widely used in other disciplines. Experiments are often unnecessarily large and surveys suggest that more than half have obvious statistical errors. Better training in experimental design would be cost effective in making better use of scarce resources, and would help to reduce the use of laboratory animals, which is an ethically desirable goal.
Similar content being viewed by others
References
Altman DG (1982a) Misuse of statistics is unethical. In: Gore SM, Altman DG (eds) Statistics in practice. British Medical Association, London, pp 1–2
Altman DG (1982b) Statistics in medical journals. Stat Med 1:59–71
Arcos JC, Argus MF, Wolf G (eds) (1968) Chemical induction of cancer. Academic Press, London
Balls M, Goldberg AM, Fentem JH et al. (1995) The three Rs: the way forward. ATLA 23:838–866
Cox DR (1958) Planning experiments. Wiley, New York
Elashoff JD (1995) nQuery advisor user’s guide. Statistical Solutions, Boston, MA
Felton RP, Gaylor DW (1989) Multistrain experiments for screening toxic substances. J Toxicol Environ Health 26:399–411
Festing MFW (1979) Inbred strains in biomedical research, 1st edn. Macmillan, London, Basingstoke
Festing MFW (1982) Genetic contamination of laboratory animal colonies: an increasingly serious problem. ILAR News 25:6–10
Festing MFW (1987) Genetic factors in toxicology: implications for toxicological screening. Crit Rev Toxicol 18:1–26
Festing MFW (1992) The scope for improving the design of laboratory animal experiments. Lab Anim 26:256–267
Festing MFW (1993) Genetic variation in outbred rats and mice and its implications for toxicological screening. J Exp Anim Sci 35:210–220
Festing MFW (1994) Reduction of animal use: experimental design and quality of experiments. Lab Anim 28:212–221
Festing MFW (1995) Use of a multi-strain assay could improve the NTP carcinogenesis bioassay program. Environ Health Perspect 103:44–52
Festing MFW (1996) Are animal experiments in toxicological research the ‘right’ size? In: Morgan BJT (ed) Statistics in toxicology. Clarendon Press, Oxford, pp 3–11
Festing MFW, Lovell DP (1995) The need for statistical analysis of rodent micronucleus test data: comment on the paper by Ashby and Tinwell. Mutat Res 329:221–224
Fisher RA (1960) The design of experiments, 7th. edn. Hafner, New York
Gartner K (1990) A third component causing random variability beside environment and gentype. A reason for limited success of a 30-year long effort to standardize laboratory animals. Lab Anim 24:71–77
Gore SM, Jones IG, Rytter EC (1977) Misuse of statistical methods: critical assessment of articles in BMJ from January to March 1976. BMJ 1:85–87
Heston WE (1968) Genetic aspects of experimental animals in cancer research. Jpn Cancer Assoc. Gann Monogr 5:3–15
Mead R (1988) The design of experiments. Cambridge University Press, Cambridge, New York
Nohynek GJ, Longeart L, Geffray B et al. (1993) Fat, frail and dying young: survival, body weight and pathology of the Charles River Sprague-Dawley derivedrat prior to and since the introduction of the VAFR variant in 1988. Hum Exp Toxicol 12:87–98
Papaioannou VE, Festing MFW (1980) Genetic drift in a stock of laboratory mice. Lab Anim 14:11–13
Sontag JM, Page NP, Saffiotti U (1976) Guidelines for carcinogen biassay in small rodents, DHERW Publication no. (NIH) 76-801 ed. National Cancer Institute, Carcinogenesis Technical Report Series no. 1, Bethesda
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Festing, M.F.W. Variation and its implications for the design of experiments in toxicological research. Comparative Haematology International 7, 202–207 (1997). https://doi.org/10.1007/BF02658690
Issue Date:
DOI: https://doi.org/10.1007/BF02658690