Challenges and opportunities of translating animal research into human trials in Ethiopia.

BACKGROUND AND OBJECTIVES: Although the goal of translational research is to bring biomedical knowledge from the laboratory to clinical trial and therapeutic products for improving health, this goal has not been well achieved as often as desired because of many barriers documented in different countries. Therefore, the aim of this study was to investigate the challenges and opportunities of translating animal research into human trials in Ethiopia. METHODS: A descriptive qualitative study, using in-depth interviews, was conducted in which preclinical and clinical trial researchers who have been involved in animal research or clinical trials as principal investigator were involved. Data were analyzed using inductive thematic process. RESULTS: Six themes were emerged for challenges: lack of financial and human capacity, inadequate infrastructure, operational obstacles and poor research governance, lack of collaboration, lack of reproducibility of results and prolonged ethical and regulatory approval processes. Furthermore, three themes were synthesized for opportunities: growing infrastructure and resources, improved human capacity and better administrative processes and initiatives for collaboration. CONCLUSION AND RECOMMENDATIONS: The study found that the identified characteristics/features are of high importance either to hurdle or enable the practice of translating animal research into human trials. The study suggests that there should be adequate infrastructure and finance, human capacity building, good research governance, improved ethical and regulatory approval process, multidisciplinary collaboration, and incentives and recognition for researchers to overcome the identified challenges and allow translating of animal research into human trials to proceed more efficiently.

What exactly is 'N' in cell culture and animal experiments?

Biologists determine experimental effects by perturbing biological entities or units. When done appropriately, independent replication of the entity-intervention pair contributes to the sample size (N) and forms the basis of statistical inference. If the wrong entity-intervention pair is chosen, an experiment cannot address the question of interest. We surveyed a random sample of published animal experiments from 2011 to 2016 where interventions were applied to parents and effects examined in the offspring, as regulatory authorities provide clear guidelines on replication with such designs. We found that only 22% of studies (95% CI = 17%-29%) replicated the correct entity-intervention pair and thus made valid statistical inferences. Nearly half of the studies (46%, 95% CI = 38%-53%) had pseudoreplication while 32% (95% CI = 26%-39%) provided insufficient information to make a judgement. Pseudoreplication artificially inflates the sample size, and thus the evidence for a scientific claim, resulting in false positives. We argue that distinguishing between biological units, experimental units, and observational units clarifies where replication should occur, describe the criteria for genuine replication, and provide concrete examples of in vitro, ex vivo, and in vivo experimental designs.

Rethinking 3R strategies: Digging deeper into AnimalTestInfo promotes transparency in in vivo biomedical research.

In the European Union (EU), animal welfare is seen as a matter of great importance. However, with respect to animal experimentation, European citizens feel quite uninformed. The European Directive 2010/63/EU for the protection of laboratory animals aims for greater transparency and requires that a comprehensible, nontechnical summary (NTS) of each authorised research project involving animals is published by the respective Member State. However, the NTSs remain sleeping beauties if their contents are not easily and systematically accessible. The German web-based NTS database AnimalTestInfo is a unique channel for scientists to communicate their work, and provides the opportunity for large-scale analyses of planned animal studies to inform researchers and the public. For an in-depth meta-analysis, we classified the duly completed NTSs submitted to AnimalTestInfo in 2014 and 2015 according to the International Classification of Diseases and Related Health Problems (ICD) system. Indexing the NTSs with ICD codes provided a fine-grained overview of the prospective uses of experimental animals. Using this approach, transparency, especially for highly controversial animal research involving, for example, nonhuman primates, is fostered, as it enables pinpointing the envisaged beneficiary down to the level of the addressed disease. Moreover, research areas with many planned projects involving animals can be specified in detail. The development of 3R (replacement, reduction, and refinement) measures in these research areas may be most efficient, as a large number of experimental animals would benefit from it. Indexing NTSs with ICD codes can support governments and funding agencies in advancing target-oriented funding of 3R research. Data drawn from NTSs can provide a basis for the development, validation, and implementation of directed 3R strategies as well as guidance for rethinking the role of animal research models.

Quality of interventional animal experiments in Chinese journals: compliance with ARRIVE guidelines.

BACKGROUND: In view of the inadequacy and incompleteness of currently-reported animal experiments and their overall poor quality, we retrospectively evaluated the reporting quality of animal experiments published in Chinese journals adhering to the Animal Research: Reporting of In Vivo Experiments (ARRIVE) guidelines. RESULTS: The databases CNKI, WanFang, VIP, and CBM were searched from inception until July 2018. Two appropriately-trained reviewers screened and extracted articles independently. The ARRIVE guidelines were used to assess the quality of the published reports of animal experiments. The compliance rate of every item was analyzed relative to their date of publication. A total of 4342 studies were included, of which 73.0% had been cited ≤5 times. Only 29.0% (1261/4342) were published in journals listed in the Chinese Science Citation Database. The results indicate that the compliance rate of approximately half of the sub-items (51.3%, 20/39) was less than 50%, of which 65.0% (13/20) was even less than 10%. CONCLUSIONS: The reporting quality of animal experiments in Chinese journals is not at a high level. Following publication of the ARRIVE guidelines in 2010, the compliance rate of the majority of its requirements has improved to some extent. However, less attention has been paid to the ethics and welfare of experimental animals, and a number of specific items in the Methods, Results, and Discussion sections continue to not be reported in sufficient detail. Therefore, it is necessary to popularize the ARRIVE guidelines, advocate researchers to adhere to them in the future, and in particular promote the use of the guidelines in specialized journals in order that the design, implementation, and reporting of animal experiments is promoted, to ultimately improve their quality.

Are the Lives of Animals Well-spent in Laboratory Science Research? A Study of Orthopaedic Animal Studies in Turkey.

BACKGROUND: As in all fields of medicine, animal studies are widely performed in orthopaedics and have increased in number over time. However, it is not clear to what extent these studies provide a basis for future research or advancements in clinical science. Concerns about the reliability and translational ability of animal studies have been reported, and major orthopaedic journals and organizations are encouraging the reduction of unnecessary experiments on animals. QUESTION/PURPOSES: (1) What proportion of animal studies conducted for orthopaedic research in Turkey were never published? And of those that were published, how long did it take to publish? (2) What proportion of those studies were published in journals with an Impact Factor of 2 or more? (3) What proportion of those published papers were never cited or cited only once? (4) What was the contribution to science of an animal euthanized for orthopaedic research in Turkey? METHODS: We reviewed all oral and poster presentations at the Turkish National Congress of Orthopaedics and Traumatology from 2009 to 2017 (retrieved from the archives of Acta Orthopaedica et Traumatologica Turcica), as well as all postgraduate theses in orthopaedics from 1991 to 2017 (retrieved from the archives of the National Thesis Center of the Council of Higher Education) to identify all orthopaedic studies that involved animals. We searched the keywords "animal studies," "experimental studies," and "orthopaedics" in these archives. We defined animal research as orthopaedic studies based on animal models. From this search and using that definition, 252 studies were identified. Of those, 4% (9) were excluded as they were thesis studies with no abstract in the archives. Thus, a total of 243 animal studies performed in Turkey were included for analysis in this retrospective study. The abstracts of these studies were examined to determine the study model (such as bone fracture models, tendon healing models, cartilage models) and number of euthanized animals. Between 1991 and 2017, 9412 vertebrate animals were euthanized for these studies. We searched PubMed, Google Scholar, ResearchGate, and ORCID to determine whether these papers were subsequently published, in which journal, and how long after the initial presentation publication occurred. The Web of Science 2019 database was used to determine the Impact Factor of the journals, the total citation count of each study, and the mean annual citation for each study (citations per year). For purposes of this analysis, we divided journals into those with an Impact Factor of 2 or more, 4 or more, and those with an Impact Factor below 2. The mean annual citation per euthanized animal (citations per animal per year) was calculated to determine the contribution of a euthanized animal to science. RESULTS: A total of 42% (101 of 243) of the animal studies in Turkey were never published. For all published studies, the mean time to publication was 2.2 ± 2.6 years (95% CI 1.7 to 2.6). The proportion of studies published in orthopaedic journals with an Impact Factor of 2 or more was 14% (34 of 243). Among the 142 published papers, 38% (54) were either never cited or were cited only once, and the mean citations per year was 1.1 ± 1.7 (95% CI 0.7 to 1.3). The mean citations per animal/year among the 142 published studies was 0.03 ± 0.04 (95% CI 0.02 to 0.04). CONCLUSION: In the 243 theses and national congress presentations, 9412 animals were euthanized. Based on the low percentage of papers using animals that were euthanized and the very low proportion of studies published in higher-Impact Factor journals or garnering more than a single citation, in aggregate, little seems to have been gained from the loss of animal life. Future studies should try to replicate or refute our results in other countries. CLINICAL RELEVANCE: Orthopaedic researchers should try to reduce their use of unnecessary animal studies, for example, by reporting on the use of the "3Rs" (replacement, reduction, and refinement) in the development of an animal study design, as well as through following universal guidelines so that a study might have a clinical impact. Researchers should not conduct an animal study until they are convinced that the expected results are quite likely to deliver substantial benefit to people or to advance science in a meaningful way; although this seems intuitive, our results suggest that this may not be taking place. Ethics committees in Turkey should consider more detailed questioning before approving animal studies. If our results are replicated elsewhere, then a broader look at how these approvals are conducted should be performed.

Publication and Citation Analysis of Medical Doctors' Residency Master's Theses Involving Animal Experiments on Rats in Turkey.

The number of non-human animals used in research has increased in line with advances in medical technology, although it has previously been shown that these experiments demonstrate poor human utility. This study aimed to determine the effectiveness of animal studies on rats that were performed as part of medical doctors' residency master's theses prepared in Turkey between January 2006 and December 2015. The number of thesis-derived published papers from each year, as well as the subsequent citation rate of these papers, was determined. Results from 34% of the 656 analysed studies (226/656) were published as papers in PubMed-indexed journals. These 226 studies got 1803 subsequent citations in total. Citation counts were statistically significantly different in 2009 and 2010, as compared to 2011, 2013, 2014 and 2015. Previous studies showed that the usual main objective for carrying out animal studies in Turkey was the preparation of a thesis or the furthering of an academic career (i.e. personal self-interest). In the current study, the publication rate and the number of subsequent citations of these thesis-derived papers were both low, and thus, the contribution of these animal studies to scientific progress is doubtful. It is recommended that institutional research ethics committees should be much more highly selective in approving the use of animals for the purposes of student thesis preparation.

Factors Related to Animal Neglect in Brazilian Laboratories.

Our objectives were to identify the prevalence of negligence of laboratory animals in Brazil, to determine the primary factors associated with its occurrence and to suggest prevention strategies. A questionnaire was made available online between October 2015 and March 2016. A total of 116 respondents with experience in the use of laboratory animals and/or the use of alternative methods answered the questionnaire. Most respondents were women (77 respondents, 66.4%), a significant proportion had a degree in Veterinary Medicine (31 respondents, 27.2%), and a majority used animals in their work (88 respondents, 75.9%). Of the 88 animal users, 23 supplied information on the numbers and species of animals they used. When asked whether they knew that Brazilian law forbade animal experimentation when alternative methods exist, seven (9.1%) respondents mentioned Act 9605/1998. Most, but not all, respondents (96 respondents, 82.8%) submitted their projects to an Animal Use and Ethics Committee (AUEC), and many (65 respondents, 56%) reported their belief that animal neglect occurred at their institution. Negligence was found to be associated with: institutions where the numbers of animals used were not recorded (p = 0.008); institutions where respondents were unaware of the relevant legislation, that is, Act 9605/1998 (p = 0.042); or where there was evidence that not all project proposals were submitted to the AUEC or evidence of no submissions at all (p = 0.022). Negligence of animals was found to be highly prevalent. Prevention strategies might involve increased transparency to the general public, the empowerment of individuals that work with animals to report any concerns, optimised inspection of facilities where animal work is carried out and significant improvements to the role of AUECs.

Experimental Animal Use in Turkey: A Comparison with Other Countries.

The legal structure that governs animal use in Turkey is in line with that of the European Union (EU). In 2004, legislation on the use of animals for experimental and other scientific purposes was established in Turkey for the first time. The present study aimed to compare the data on experimental animal use in Turkey (during the period 2008-2017) with similar reports from selected countries (the United States, Australia, Canada and the EU). In Turkey, a total of 2,104,828 animals were used for experimental and other scientific purposes during 2008-2017. Of the animals used, 758,887 were fish (36%), 433,417 rats (21%), 302,512 birds other than quail (14%) and 285,531 mice (13%). According to a breakdown by purpose for use, in Turkey during 2009-2017, out of a total number of 1,955,307 animals used, 56% were for fundamental biological studies, with a high proportion used for research on animal disease. Compared with the other countries, fewer animals were used in Turkey although the national trend seems to indicate that the number is fluctuating. Further studies are required to uncover the reasons behind this reduced animal use in Turkey, as compared to other countries.

An Estimate of the Number of Animals Used for Scientific Purposes Worldwide in 2015.

Few attempts have been made to estimate the global use of animals in experiments, since our own estimated figure of 115.2 million animals for the year 2005. Here, we provide an update for the year 2015. Data from 37 countries that publish national statistics were standardised against the definitions of 'animals' and 'procedures' used in the European Union (EU) Directive 2010/63/EU. We also applied a prediction model, based on publication rates, to estimate animal use in a further 142 countries. This yielded an overall estimate of global animal use in scientific procedures of 79.9 million animals, a 36.9% increase on the equivalent estimated figure for 2005, of 58.3 million animals. We further extrapolated this estimate to obtain a more comprehensive final global figure for the number of animals used for scientific purposes in 2015, of 192.1 million. This figure included animals killed for their tissues, normal and genetically modified (GM) animals without a harmful genetic mutation that are used to maintain GM strains and animals bred for laboratory use but not used. Since the 2005 study, there has been no evident increase in the number of countries publishing data on the numbers of animals used in experiments. Without regular, accurate statistics, the impact of efforts to replace, reduce and refine animal experiments cannot be effectively monitored.

Conscientious Objection to Animal Testing: A Preliminary Survey Among Italian Medical and Veterinary Students.

The use of animals for educational and research purposes is common in both veterinary and human medicine degree courses, and one that involves important ethical considerations. The aim of this study was to assess the extent of differences between the knowledge and attitudes of veterinary students and medical students on animal bioethics, on alternative strategies and on their right to conscientiously object to animal experimentation. To this end, a questionnaire was completed by 733 students (384 human medicine students (HMS) and 349 veterinary medicine students (VMS)). VMS were more aware than HMS (72.2% and 59.6%, respectively) of the existence of an Italian law on the right to conscientiously object to animal experimentation. However, very few of them had exercised this right. Many VMS (43.3%) felt that animal bioethics courses should be mandatory (only 17.4% of HMS felt the same way). More VMS than HMS (81.7% and 59.1%, respectively) expressed an interest in attending a course on alternatives to animal experimentation. The data suggest the need for appropriate educational interventions, in order to allow students to make choices based on ethical principles. Fostering close collaborations between departments of human medicine and veterinary medicine, for example, through shared study modules, could promote the development of ethical competence as a basic skill of students of both veterinary and human medicine courses.

Applying parametric models to survival data: tradeoffs between statistical significance, biological plausibility, and common sense.

Parametric models for survival data help to differentiate aging from other lifespan determinants. However, such inferences suffer from small sizes of experimental animal samples and variable animals handling by different labs. We analyzed control data from a single laboratory where interventions in murine lifespan were studied over decades. The minimal Gompertz model (GM) was found to perform best with most murine strains. However, when several control datasets related to a particular strain are fitted to GM, strikingly rigid interdependencies between GM parameters emerge, consistent with the Strehler-Mildvan correlation (SMC). SMC emerges even when survival patterns do not conform to GM, as with cancer-prone HER2/neu mice, which die at a log-normally distributed age. Numerical experiments show that SMC includes an artifact whose magnitude depends on dataset deviation from conformance to GM irrespectively of the noisiness of small datasets, another contributor to SMC. Still another contributor to SMC is the compensation effect of mortality (CEM): a real tradeoff between the physiological factors responsible for initial vitality and the rate of its decline. To avoid misinterpretations, we advise checking experimental results against a SMC based on historical controls or on subgroups obtained by randomization of control animals. An apparent acceleration of aging associated with a decrease in the initial mortality is invalid if it is not greater than SMC suggests. This approach applied to published data suggests that the effects of calorie restriction and of drugs believed to mimic it are different. SMC and CEM relevance to human survival patterns is discussed.

Two years later: journals are not yet enforcing the ARRIVE guidelines on reporting standards for pre-clinical animal studies.

There is growing concern that poor experimental design and lack of transparent reporting contribute to the frequent failure of pre-clinical animal studies to translate into treatments for human disease. In 2010, the Animal Research: Reporting of In Vivo Experiments (ARRIVE) guidelines were introduced to help improve reporting standards. They were published in PLOS Biology and endorsed by funding agencies and publishers and their journals, including PLOS, Nature research journals, and other top-tier journals. Yet our analysis of papers published in PLOS and Nature journals indicates that there has been very little improvement in reporting standards since then. This suggests that authors, referees, and editors generally are ignoring guidelines, and the editorial endorsement is yet to be effectively implemented.

Sample size estimation for pilot animal experiments by using a Markov Chain Monte Carlo approach.

The statistical determination of sample size is mandatory when planning animal experiments, but it is usually difficult to implement appropriately. The main reason is that prior information is hardly ever available, so the assumptions made cannot be verified reliably. This is especially true for pilot experiments. Statistical simulation might help in these situations. We used a Markov Chain Monte Carlo (MCMC) approach to verify the pragmatic assumptions made on different distribution parameters used for power and sample size calculations in animal experiments. Binomial and normal distributions, which are the most frequent distributions in practice, were simulated for categorical and continuous endpoints, respectively. The simulations showed that the common practice of using five or six animals per group for continuous endpoints is reasonable. Even in the case of small effect sizes, the statistical power would be sufficiently large (≥ 80%). For categorical outcomes, group sizes should never be under eight animals, otherwise a sufficient statistical power cannot be guaranteed. This applies even in the case of large effects. The MCMC approach demonstrated to be a useful method for calculating sample size in animal studies that lack prior data. Of course, the simulation results particularly depend on the assumptions made with regard to the distributional properties and effects to be detected, but the same also holds in situations where prior data are available. MCMC is therefore a promising approach toward the more informed planning of pilot research experiments involving the use of animals.

The use of gatekeeping procedures in the statistical planning of animal experiments.

Statistical sample size calculation is essential when planning animal experiments in basic medical research. Usually, such trials involve the testing of multiple hypotheses, and interpreting them in a confirmative manner would require the appropriate adjustment of the Type 1 error. This has to be taken into account as early as possible during sample size estimation - otherwise, all the results obtained would be exploratory, i.e. without cogency. In this paper, the concept of gatekeeping is introduced, along with alternative approaches for Type 1 error adjustment. The application of gatekeeping to the calculation of sample size is demonstrated by using data sets from case studies. Overall, the evaluation of these examples showed that gatekeeping is able to keep the required number of animals comparatively small. In contrast to exploratory planning, which led to the lowest sample sizes, gatekeeping suggested a mean increase of 12% in sample size, while conservative Bonferroni adjustment raised the sample size by 34% on average. Gatekeeping is a prominent strategy for handling the multiple testing problem, and has been proven to keep the required sample sizes in animal studies comparatively low. Therefore, it is a suitable approach to a compromise between the Three Rs principle of reduction and the appropriate handling of the multiplicity issue in animal trials with a confirmative focus.

Statistics of Scientific Procedures on Living Animals 2014: A new format, and hopefully a new era of diminishing animal experimentation?

The Annual Statistics of Scientific Procedures on Living Animals Great Britain 2014 reports a welcome decline in animal experimentation in the UK. However, caution has to be exercised when interpreting these most recent figures, due to the significant changes made to satisfy the requirements of Directive 2010/63/EU as to what information is reported and how it is reported. Comparisons to the figures and trends reported in previous years is difficult, so this paper focuses on the specifics of the current report, providing information on overall animal use and highlighting specific issues associated with genetically-altered animals, fish and primates. There is a detailed discussion of the extent of the changes, commenting on the benefits and disadvantages of the new format, in areas such as severity of procedures, legislation and techniques of special interest. It also considers the consequences of the changes on the effective monitoring of laboratory animal use, the openness and transparency regarding the impacts of animal use, and the implementation of Three Rs initiatives. In addition, suggestions for further improvements to the new format are made to the Home Office.

Statistics of Scientific Procedures on Living Animals Great Britain 2015 - highlighting an ongoing upward trend in animal use and missed opportunities.

The Annual Statistics of Scientific Procedures on Living Animals Great Britain 2015 indicate that the Home Office were correct in recommending that caution should be exercised when interpreting the 2014 data as an apparent decline in animal experiments. The 2015 report shows that, as the changes to the format of the annual statistics have become more familiar and less problematic, there has been a re-emergence of the upward trend in animal research and testing in Great Britain. The 2015 statistics report an increase in animal procedures (up to 4,142,631) and in the number of animals used (up to 4,069,349). This represents 1% more than the totals in 2013, and a 7% increase on the procedures reported in 2014. This paper details an analysis of these most recent statistics, providing information on overall animal use and highlighting specific issues associated with genetically-altered animals, dogs and primates. It also reflects on areas of the new format that have previously been highlighted as being problematic, and concludes with a discussion about the use of animals in regulatory research and testing, and how there are significant missed opportunities for replacing some of the animal-based tests in this area.

Statistics of Scientific Procedures on Living Animals 2013: Experimentation continues to rise--the reliance on genetically-altered animals must be addressed.

The 2013 Statistics of Scientific Procedures on Living Animals reveal that the level of animal experimentation in Great Britain continues to rise, with 4.12 million procedures being conducted. The figures indicate that this is almost exclusively a result of the breeding and use of genetically-altered (GA) animals (i.e. genetically-modified animals, plus those with harmful genetic defects). The breeding of GA animals increased to over half (51%) of all the procedures, and GA animals were involved in 61% of all the procedures. Indeed, if these animals were removed from the statistics, the number of procedures would actually have declined by 4%. It is argued that the Coalition Government has failed to address this issue, and, as a consequence, will not be able to deliver its pledge to reduce animal use in science. Recent publications supporting the need to reassess the dominance of genetic alteration are also discussed, as well as the need to move away from the use of dogs as the default second species in safety testing. The general trends in the species used, and the numbers and types of procedures, are also reviewed. Finally, forthcoming changes to the statistics are discussed.

Half the price, twice the gain: How to simultaneously decrease animal numbers and increase precision with good experimental design.

Animal research often involves experiments in which the effect of several factors on a particular outcome is of scientific interest. Many researchers approach such experiments by varying just one factor at a time. As a consequence, they design and analyze the experiments based on a pairwise comparison between two groups. However, this approach uses unreasonably large numbers of animals and leads to severe limitations in terms of the research questions that can be answered. Factorial designs and analyses offer a more efficient way to perform and assess experiments with multiple factors of interest. We will illustrate the basic principles behind these designs, discussing a simple example with only two factors before suggesting how to design and analyze more complex experiments involving larger numbers of factors based on multiway analysis of variance.

Incorporating sources of correlation between outcomes: An introduction to mixed models.

Animal research often involves measuring the outcomes of interest multiple times on the same animal, whether over time or for different exposures. These repeated outcomes measured on the same animal are correlated due to animal-specific characteristics. While this repeated measures data can address more complex research questions than single-outcome data, the statistical analysis must take into account the study design resulting in correlated outcomes, which violate the independence assumption of standard statistical methods (e.g. a two-sample t-test, linear regression). When standard statistical methods are incorrectly used to analyze correlated outcome data, the statistical inference (i.e. confidence intervals and p-values) will be incorrect, with some settings leading to null findings too often and others producing statistically significant findings despite no support for this in the data. Instead, researchers can leverage approaches designed specifically for correlated outcomes. In this article, we discuss common study designs that lead to correlated outcome data, motivate the intuition about the impact of improperly analyzing correlated outcomes using methods for independent data, and introduce approaches that properly leverage correlated outcome data.

Heterogeneity of animal experiments and how to deal with it.

Heterogeneity of study samples is ubiquitous in animal experiments. Here, we discuss the different options of how to deal with heterogeneity in the statistical analysis of a single experiment. Specifically, data from different sub-groups (e.g. sex, strain, age cohorts) may be analysed separately, heterogenization factors may be ignored and data pooled for analysis, or heterogenization factors may be included as additional variables in the statistical model. The cost of ignoring a heterogenization factor is an inflated estimate of the variance and a consequent loss of statistical power. Therefore, it is usually preferable to include the heterogenization factor in the statistical model, especially if the heterogenization factor has been introduced intentionally (e.g. using both sexes). If heterogenization factors are included, they can be treated either as fixed factors in an analysis of variance design or sometimes as random effects in mixed effects regression models. Finally, for an appropriate sample size estimation, it is necessary to decide whether to treat heterogenization factors as nuisance variables, or whether the experiment should be powered to be able to detect not only the main effect of the treatment but also interactions between heterogenization factors and the treatment variable.