Methods of scientific research

[reading time: 7 minutes]

The positivist scientific paradigm currently in use dates back to the Galilean method: the researcher proceeds from the collection of data from observation and experimentation, on these bases he or she develops hypotheses and theories. In a nutshell, it can be said that this interpretation of reality is based on models susceptible to improvement and modification as new facts and events occur.

At the moment, the implementation of a scientific study must respect the norms of Good Clinical Practice GCP, an international standard of scientific ethics and quality that regulates the design, conduct, reporting and publication of clinical trials involving human subjects. These standards not only do publicly guarantee the protection of rights, safety and wellbeing of the subjects participating in the study, in accordance with the principles established by the Helsinki Declaration, but also the reliability of the data related to the study. The GCP was developed by the International Council for Harmonisation (IHC) in 1996, adopted by the European Union in 1996 and implemented for clinical trials from 17 January 1997 also in the UK. Nevertheless, the GCP guidelines were not enforced by law at that time in the UK but in 2004 with The Medicines for Human Use (Clinical Trial) Regulation.¹

A first revision of the GCP was issued in 2016 and another one is expected in 2022. The 2016 update, ICH E6(R2), came from the need to respond to the ongoing changes in conduct and management of the clinical trials, in order to provide some guidelines integrated and validated within an ecosystem that becomes more and more digitized, in which clinical experimentation employs designs of increasing complexity.

Before starting any kind of study involving human beings it is necessary to obtain the approval of the Research Ethics Committee. These institutions, scattered over the territory, are independent bodies that have the responsibility to provide a public guarantee for the protection of the rights, safety and wellbeing of the subjects participating in clinical studies, by expressing an ethical and scientific opinion. In the UK this role is acted by more than 80 NHS Research Ethics Committees, each of which is composed of 15 members who are not directly involved in the Research field or have any interest in it. What’s more, they are not registered healthcare workers. On the other hand, a government agency, the Medicines and Healthcare Product and Regulatory Agency (MHRA) is the regulatory authority responsible for reviewing, authorizing and overseeing clinical trial in the UK.

Moreover, since 2005 whoever intends to carry out a clinical pharmacologic experimentation has to register the study in one of the official databases available online. This is essential for transparency as well as for the reduction of publication bias. In fact, it is important to have news of studies that have been started and then suspended, or of studies whose results have not been published because they were not considered “positive”, etc. The failure to publish these studies may, in fact, not only translate into the loss of information for the scientific community but also in a duplication of the same study in later times, entailing a waste of resources and an unnecessary involvement of patients in trials.

The registration has to take place on a date prior to the recruiting of the first participant in the study, and allows the allocation of a unique code that is used to identify the study and that is also reported on the journal publishing the study. However, it is to be noted that at the moment not all journals request the identification code and not all studies are experimental, that is, subject to this obligation.

The WHO has established an International Clinical Trial Registry Platform (ICTRP) in order to allow all those involved in taking health care decisions to access a complete visualization of the research studies. One page of its website lists the main registries into which the studies can be registered. For example, in Europe, EU Clinical Trials Register website, holds the register of the clinical trials at EudraCT and provides a page that allows a search by protocol and by results. For clinical trials (other than adult Phase 1 studies) involving both UK and EU sites, a record in the EU Clinical Trials Register will satisfy the Research Database Conditions of REC’s (Research Ethics Committee) opinion in favor of registration. Also clinical trials carried out outside the EU / EEA, connected to European paediatric-medicine development can be registered here. In all other cases, as a result of Brexit, the EU CTR does not apply in the United Kingdom (UK). Therefore new rules on clinical trials are needed. ISRCTN is the UK’s preferred clinical trials registry.
The platform clinicaltrials.gov, made available by the American National Library of Medicine, is also extremely relevant. This is a database of publicly and privately funded studies conducted all over the world (featuring more than 380.000 studies in 220 countries).

Once fulfilled the ethical and bureaucratic obligations a researcher intending to conduct a study, first of all, has to establish if the research question concerns numerical and statistical studies or more descriptive studies, perhaps linked to the motivations and attitude of the people involved. In fact, science investigations are defined as quantitative studies that try to establish a causality link between an intervention (eg, a drug) and an outcome (eg, the alleviation of pain in a patient). The other types of science investigation are defined as qualitative studies , based on interpretivist paradigms that aim to understand reality and not to explain it by measuring it.

There are also mixed approaches, which propose an integrated use of qualitative and quantitative methods to understand the most complex and multidimensional aspects of clinical practice.

In principles, the implementation of a scientific study can be articulated into three phases:

• Choice of the research design: a phase in which the research question is set, the hypotheses are formulated and the practical organization of the study is planned. Fundamental in this regard is the writing of the protocol, which will include several parameters, like the title of the study, the primary objective, the size of the population and any other detail required for the type of study selected. In order to write the protocol in accordance with the rules accepted by the scientific community, the researcher may use various guidelines explaining which elements are to be included and which are different depending on the study. By way of example we mention here the SPIRIT guidelines, containing the recommendations to define the standard elements of the protocol for the clinical trials and the PRISMA-P guidelines to be used for the writing of protocols for systematic reviews;
• data collection and analysis: the procedures for the quantitative and qualitative research are different. In the first case in fact we start from a representative sample of the population, while in the second case it is very important to understand the peculiarities that distinguish every single participant. The data collected are considered soft when they are subjective and complex, or hard when they are standardized and suitable for analysis with mathematical tools.
• interpretation/presentation of the results: the structural differences between the two approaches is reflected also in the results. Quantitative research scrutinizes its data through statistical analysis in the attempt to search for correlations and causal interrelations between variables, whereas qualitative research typically reports narrations or quotes to illustrate the thoughts of the subjects studied.

When conducting studies based on the scientific method, one of the researchers’ main concerns is to minimize the risk of error. This is an extremely important matter, which gave rise to the development of ever more scrupulous and rigorous methodologies that must be followed in order to avoid mistakes.

Errors can be relatively trivial and naive, such as those that link a random association to causality: If, for example, a statistical analysis shows a positive correlation between exercise and skin cancer, this does not mean that gymnastics causes cancer but that there is probably a third factor (perhaps the increased exposure to the sun) that had not been taken into account.

Nevertheless, there are many other factors generating inaccuracy in experimental studies. These are error sources that need to be identified and defined in order to manage them. They Include different kinds of bias, ie, systematic errors or deviations (distortions) from reality in the results or in the inference.

Lastly, we need to take under consideration the placebo and nocebo effects, which are difficult to measure objectively and may affect the outcome of a study. The placebo effect is a response activated by mechanisms related to expectations and conditioning, and thanks to which the patient gains benefits, sometimes comparable to those produced by a medicine although without having actually taken it. On the other hand, the nocebo effect, only studied in more recent times, refers to the increase of the perceived pain in a patient exposed to verbal suggestions indicating the risk of a worsening³.

Bibliography

1. The Medicines for Human Use (Clinical Trial) Regulation 2004
2. Müller AR, Brands MMMG, van de Ven PM, Roes KCB, Cornel MC, van Karnebeek CDM, Wijburg FA, Daams JG, Boot E, van Eeghen AM. Systematic Review of N-of-1 Studies in Rare Genetic Neurodevelopmental Disorders: The Power of 1. Neurology. 2021 Mar 16;96(11):529.
3. P. Enck, Benedetti F., New Insight into the Placebo and Nocebo Responses. Neuron, Volume 59, Issue 2, 31 July 2008, Pages 195-206, https://www.sciencedirect.com/science/article/pii/S0896627308005850