Basket | Login | Register


Clinical trials

Ethics of dementia research

What are clinical trials and how are they controlled/governed?

A clinical trial is a biomedical/health-related study into the effects on humans of a new medical treatment (medicine/drug, medical device, vaccine or new therapy), sometimes called an investigational medicinal product (IMP). Before a new drug is authorised and can be marketed, it must pass through several phases of development including trial phases in which its safety, efficacy, risks, optimal use and/or benefits are tested on human beings. Existing drugs must also undergo clinical testing before they can be used to treat other conditions than that for which they were originally intended.

Organisations conducting clinical trials in the European Union must, if they wish to obtain marketing authorisation, respect the requirements for the conduct of clinical trials. These can be found in the Clinical Trials Directive (“Directive 2001/20/EC of the European Parliament and of the Council of 4 April 2001 on the approximation of the laws, regulations and administrative provisions of the Member States relating to the implementation of good clinical practice in the conduct of clinical trials on medicinal products for human use”).

There are also guidelines to ensure that clinical trials are carried out in accordance with good clinical practice. These are contained in the “Commission Directive 2005/28/EC of 8 April 2005 laying down principles and detailed guidelines for good clinical practice as regards investigational medicinal products for human use, as well as the requirements for authorisation of the manufacturing or importation of such products” (also known as the Good Clinical Practice or GCP for short). This document provides more concrete guidelines and lends further support to the Clinical Trials Directive.

The London-based European Medicines Agency (EMA) has published additional, more specific guidelines which must also be respected. These include guidelines on inspection procedures and requirements related to quality, safety and efficacy.

Copies of the above-mentioned documents in 22 languages can be found at:

The protection of people participating in clinical trials (and in most cases in other types of research) is further promoted by provisions of:

  • the European Convention on Human Rights and Biomedicine (Oviedo Convention, Act 2619/1998),
  • the Additional protocol to the Oviedo Convention concerning Biomedical Research
  • the Nuremberg Code of 1949,
  • the revised Helsinki Declaration of the World Medical Association regarding Ethical Principles for Medical Research Involving Human Subjects,
  • The Belmont Report  of 18 April 1979 on the Ethical Principles and Guidelines for the Protection of Human Subjects of Research.

What are the different phases of trials?

Testing an experimental drug or medical procedure is usually an extremely lengthy process, sometimes lasting several years. The overall procedure is divided into a series of stages (known as phases) which are described below.

Pre-clinical testing

Clinical testing on humans can only begin after a pre-clinical phase, involving laboratory studies (in vitro) and tests on animals, which has shown that the experimental drug is considered safe and effective.

Whilst a certain amount of testing can be carried out by means of computer modelling and by isolating cells and tissue, it becomes necessary at some point in time to test the drug on a living creature. Animal testing is an obligatory stage in the process of obtaining regulatory approval for new drugs and medicines, and hence a legal requirement (EU Directive 2001/83/EC relating to Medicinal Products for Human Use).  The necessity of carrying out prior testing on animals is also stated in the World Medical Association’s “Ethical Principles for Medical Research Involving Human Subjects.

In order to protect the well-being of research animals, researchers are guided by three principles which are called the 3Rs:

Reduce the number of animals used to a minimum

Refine the way that experiments are carried out so that the effect on the animal is minimised and animal welfare is improved

Replace animal experiments with alternative (non-animal) techniques wherever possible.

In addition, most countries will have official regulatory bodies which control animal research. Most animals involved in research are mice. However, no animal is sufficiently similar to humans (even genetically modified ones) to make human testing unnecessary. For this reason, the experimental drug must also be tested on humans.

The main phases of clinical trials

Clinical trials on humans can be divided into three main phases (literally, phase I, II and III).  Each phase has specific objectives (please see below) and the number of people involved increases as the trial progresses from one phase to the next.

Phase I trials

Phase 1 trials are usually the first step in testing a new drug or treatment on humans after successful laboratory and animal testing.  They are usually quite small scale and usually involve healthy subjects or sub-groups of patients who share a particular characteristic. The aims of these trials are:

  • to assess the safety of experimental drugs,
  • to evaluate any possible side effects,
  • to determine a safe dose range,
  • to see how the body reacts to the drug (how it is absorbed, distributed and eliminated from the body, the effects that it has on the body and the effects it has on biomarkers).

Dose ranging, sometimes called dose escalation, studies may be used as a means to determine the most appropriate dosage, but the doses administered to the subjects should only be a fraction of those which were found to cause harm to animals in the pre-clinical studies.

The process of determining an optimal dose in phase I involves quite a high degree of risk because this is the first time that the experimental treatment or drug has been administered to humans. Moreover, healthy people’s reactions to drugs may be different to those of the target patient group. For this reason, drugs which are considered to have a potentially high toxicity are usually tested on people from the target patient group.

There are a few sequential approaches to phase I trials e.g. single ascending dose studies, multiple ascending dose studies and food effect.

In single ascending dose studies (SAD), a small group of subjects receive a very low dose of the experimental drug and are then observed in order to see whether that dose results in side effects. For this reason, trials are usually conducted in hospital settings. If no adverse side effects are observed, a second group of subjects are given a slightly higher dose of the same drug and also monitored for side-effects. This process is repeated until a dose is reached which results in intolerable side effects. This is defined as the maximum tolerated dose (MTD). 

Multiple ascending dose studies (MAD) are designed to test the pharmacokinetics and pharmacodynamics of multiple doses of the experimental drug. A group of subjects receives multiple doses of the drug, starting at the lowest dose and working up to a pre-determined level. At various times during the period of administration of the drug, and particularly whenever the dose is increased, samples of blood and other bodily fluids are taken. These samples are analysed in order to determine how the drug is processed within the body and how well it is tolerated by the body.

Food effect studies are investigations into the effect of food intake on the absorption of the drug into the body. This involves two groups of subjects being given the same dose of the experimental drug but for one of the groups when fasting and for the other after a meal. Alternatively, this could be done in a cross-over design whereby both groups receive the experimental drug in both conditions in sequence (e.g. when fasting and on another occasion after a meal). Food effect studies allow researchers to see whether eating before the drug is given has any effect on the absorption of the drug by the body.

Phase II trials

Having demonstrated the initial safety of the drug (often on a relatively small sample of healthy individuals), phase II clinical trials can begin. Phase II studies are designed to explore the therapeutic efficacy of a treatment or drug in people who have the condition that the drug is intended to treat. They are sometimes called therapeutic exploratory trials and tend to be larger scale than Phase I trials.

Phase II trials can be divided into Phase IIA and Phase IIB although sometimes they are combined.

Phase IIA is designed to assess dosing requirements i.e. how much of the drug should patients receive and up to what dose is considered safe? The safety assessments carried out in Phase I can be repeated on a larger subject group. As more subjects are involved, some may experience side effects which none of the subjects in the Phase I experienced. The researchers aim to find out more about safety, side effects and how to manage them.

Phase IIB studies focus on the efficacy of the drug i.e. how well it works at the prescribed doses. Researchers may also be interested in finding out which types of a specific disease or condition would be most suitable for treatment.

Phase II trials can be randomised clinical trials which involve one group of subjects being given the experimental drug and others receiving a placebo and/or standard treatment. Alternatively, they may be case series which means that the drug’s safety and efficacy is tested in a selected group of patients. If the researchers have adequately demonstrated that the experimental drug (or device) is effective against the condition for which it is being tested, they can proceed to Phase III. 

Phase III trials

Phase III trials are the last stage before clinical approval for a new drug or device. By this stage, there will be convincing evidence of the safety of the drug or device and its efficacy in treating people who have the condition for which it was developed. Such studies are carried out on a much larger scale than for the two previous phases and are often multinational. Several years may have passed since the original laboratory and animal testing.

The main aims of Phase III trials are:

to demonstrate that the treatment or drug is safe and effective for use in patients in the target group (i.e. in people for whom it is intended)

to monitor side effects

to test different doses or different ways of administering the drug

to determine whether the drug could be used at different stages of the disease.

to provide sufficient information as a basis for marketing approval

Researchers may also be interested in showing that the experimental drug works for additional groups of people with conditions other than that for which the drug was initially developed. For example, they may be interested in testing a drug for inflammation on people with Alzheimer’s disease. The drug would have already have proven safe and obtained marketing approval but for a different condition, hence the need for additional clinical testing.

Open label extension trails

Open label extension studies are often carried out immediately after a double blind randomised clinical trial of an unlicensed drug. The aim of the extended study is to determine the safety and tolerability of the experimental drug over a longer period of time, which is generally longer than the initial trial and may extend up until the drug is licensed. Participants all receive the experimental drug irrespective of which arm of the previous trial they were in. Consequently, the study is no longer blind in that everybody knows that each participant is receiving the experimental drug but the participants and researchers still do not know which group participants were in during the initial trial.

Post-marketing surveillance studies (phase IV)

After the three phases of clinical testing and after the treatment has been approved for marketing, there may be a fourth phase to study the long-term effects of drugs or treatment or to study the impact of another factor in combination with the treatment (e.g. whether a particular drug reduces agitation).

Usually, such trials are sponsored by pharmaceutical companies and described as pharmacovigilance. They are not as common as the other types of trials (as they are not necessary for marketing permission). However, in some cases, the EMA grants restricted or provisional marketing authorisation, which is dependent on additional phase IV trails being conducted.

Expanded access to a trial

Sometimes, a person might be likely to benefit from a drug which is at various stages of testing but does not fulfil the conditions necessary for participation in the trial (e.g. s/he may have other health problems). In such cases and if the person has a life-threatening or serious condition for which there is no effective treatment, s/he may benefit from “expanded access” use of the drug. There must, however, be evidence that the drug under investigation has some likelihood of being effective for that patient and that taking it would not constitute an unreasonable risk.

The use of placebo and other forms of comparison

The main purpose of clinical drug studies is to distinguish the effect of the trial drug from other influences such as spontaneous change in the course of the disease, placebo effect, or biased observation. A valid comparison must be made with a control. The American Food and Drugs Administration recognises different types of control namely,


  • placebo,
  • active treatment with a known effective therapy or
  • dose-comparison,
  • no treatment,
  • historical treatment (which could be an adequately documented natural history of the disease or condition, or the results of active treatment in comparable patients or populations).


The EMA considers three-armed trials (including the experimental medicine, a placebo and an active control) as a scientific gold standard and that there are multiple reasons to support their use in drug development [1] .

Participants in clinical trials are usually divided into two or more groups. One group receives the active treatment with the experimental substance and the other group receives a placebo, a different drug or another intervention. The active treatment is expected to have a positive curative effect whereas the placebo is expected to have zero effect. With regard to the aim to develop more effective treatments, there are two possibilities:

1. the experimental substance is more effective than the current treatment or

2. it is more effective than no treatment at all.

According to article 11 of the International Ethical Guidelines for Biomedical Research (IEGBR) of 2002, participants allocated to the control group in a trial for a diagnostic, therapeutic or preventive intervention should receive an established effective intervention but it may in some circumstances be considered ethically acceptable to use a placebo (i.e. no treatment). In article 11 of the IEGBR, reasons for the use of placebo are:

1. that there is no established intervention

2. that withholding an established effective intervention would expose subjects to, at most, temporary discomfort or delay in relief of symptoms

3. that use of an established effective intervention as comparator would not yield scientifically reliable results and use of placebo would not add any risk of serious or irreversible harm to the subjects.

[1] November 2010, EMA/759784/2010 Committee for Medicinal Products for Human Use

The use of placebo and the issue of irreversible harm

It has been suggested that clinical trials are only acceptable in ethical terms if there is uncertainty within the medical community as to which treatment is most suitable to cure or treat a disease (National Bioethics Commission of Greece, 2005). In the case of dementia, whilst there is no cure, there are a few drugs for the symptomatic treatment of dementia. Consequently, one could ask whether it is ethical to deprive a group of participants of treatment which would have most likely improved their condition for the purpose of testing a potentially better drug (National Bioethics Commission of Greece, 2005). Can they be expected to sacrifice their own best interests for those of other people in the future? It is also important to ask whether not taking an established effective intervention is likely to result in serious or irreversible harm.

In the 2008 amended version of the Helsinki Declaration [1] (World Medical Association, 1964), the possible legitimate use of placebo and the need to protect subjects from harm are addressed.

“32. The benefits, risks, burdens and effectiveness of a new intervention must be tested against those of the best current proven intervention, except in the following circumstances:

The use of placebo, or no treatment, is acceptable in studies where no current proven intervention exists; or

               Where for compelling and scientifically sound methodological reasons the use of placebo is necessary to determine the efficacy or safety of an intervention and the patients who receive placebo or no treatment will not be subject to any risk of serious or irreversible harm. Extreme care must be taken to avoid abuse of this option.” (WMA, 1964 with amendments up to 2008)

The above is also quite similar to the position supported by the Presidential Commission for the Study of Bioethical Issues (PCSBI) (2011). In its recently published report entitled “Moral science: protecting participants in human subjects research [2] ”, the Presidential Commission argues largely in favour of a “middle ground” for ethical research, citing the work of Emanuel and Miller (2001) who state:

“A placebo-controlled trial can sometimes be considered ethical if certain methodological and ethical standards are met. It these standards cannot be met, then the use of placebos in a clinical trial is unethical.” (Emanuel and Miller, 2001 cited in PCSBI, 2011, p. 89).

One of the standards mentioned is the condition that withholding proven effective treatment will not cause more than minimal harm.

The importance of placebo groups for drug development

The ethical necessity to include a placebo arm in a clinical trial may differ depending on the type of drug being developed and whether other comparable drugs exist. For example, a placebo arm would be absolutely necessary in the testing of a new compound for which no drug has yet been developed. This would be combined with comparative arms involving other alternative drugs which have already been proven effective. For studies involving the development of a drug based on an existing compound, a comparative trial would be necessary but not necessarily with a placebo arm, or at least with a smaller placebo arm Nevertheless, the EMA emphasises the value of placebo-controlled trials in the development of new medicinal products even in cases where a proven effective drug exists:

“forbiddingplacebo-controlled trials in therapeutic areas where there are proven,  therapeutic methods would preclude obtaining reliable scientific evidence for the evaluation of new medicinal products, and be contrary to public health interest as there is a need for both new products and alternatives to existing medicinal products.” (EMA, 2001).

In 2001, concerns were raised about the interpretation of paragraph 29 of the 2000 version of the Helsinki Declaration in which prudence was called for in the use of placebo in research trials and it was advised that placebo should only be used in cases where there was no proven therapy for the condition under investigation. A document clarifying the position of the WMA regarding the use of placebo was issued by the WMA in 2001 in which it was made clear that the use of placebo might be ethically acceptable even if proven therapy was available. The current version of this statement is article 32 of the 2008 revised Helsinki Declaration (quoted in sub-section 7.2.1).

The PCSBI (2011) highlight the importance of ensuring that the design of clinical trials enables the researchers to resolve controversy and uncertainty over the merits of the trial drug and whether the trial drug is better than an existing drug if there is one. They suggest that studies which cannot resolve such questions or uncertainty are likely to be ignored by the scientific community and this would be unethical as it would mean that people had been unnecessarily exposed to risk without there being any social benefit.

Reasons for participation

People with dementia who take part in clinical trials may do so for a variety of reasons. One possible reason is that they hope to receive some form of treatment that will improve their condition or even result in a cure. This is sometimes called the “therapeutic misconception”. In such cases, clinical trials may seem unethical in that advantage is being taken of the vulnerability of some of the participants. On the other hand, the possibility of participating in such a trial may help foster hope which may even enable a person to maintain their morale.

A review of 61 studies on attitudes to trials has shed some light on why people participate in clinical trials (Edwards, Lilford and Hewison, 1998). In this review, it was found that over 60% of participants in seven studies stated that they did or would participate in clinical trials for altruistic reasons.  However, in 4 studies, over 70% of people stated that they participated out of self-interest and in two studies over 50% of people stated that they would participate in such a study out of self-interest. As far as informed consent is concerned, in two studies (which were also part of this review) 47% of responding doctors thought that few patients were actually aware that they were taking part in a clinical trial. On the other hand, an audit of four further studies revealed that at least 80% of participants felt that they had made an autonomous decision. There is no proof whether such perceptions were accurate or not. The authors conclude that self-interest was more common than altruism amongst the reasons given for participating in clinical trials but draw attention to the poor quality of some of the studies reviewed thereby suggesting the need for further research. It should not be necessary for people to justify why they are willing to participate in clinical trials. Reasons for participating in research are further discussed in section 3.2.4 insofar as they relate to end-of-life research. 

In a series of focus groups organised in 8 European countries plus Israel and covering six conditions including dementia, helping others was seen as the main reason why people wanted to take part in clinical trials (Bartlam et al., 2010). In a US trial of anti-inflammatory medication in Alzheimer’s disease in which 402 people were considered eligible, of the 359 who accepted, their main reasons for wanting to participate were altruism, personal benefit and family history of Alzheimer’s disease.

Random assignment to study groups

As people are randomly assigned to the placebo or the active treatment group, everyone has an equal chance of receiving the active ingredient or whichever other control groups are included in the study. There are possible advantages and drawbacks to being in each group and people are likely to have preferences for being a particular study group but randomization means that allocation is not in any way linked to the best interests of each participant from a medical perspective. This is not an ethical issue provided that each participant fully understands that the purpose of research is not to provide a tailor-made response to an individual’s medical condition and that while some participants benefit from participation, others do not.

There are, however, medical issues to consider. In the case in double-blind studies, neither the participant nor the investigator knows to which groups a participant has been allocated. Consequently, if a participant encounters medical problems during the study, it is not immediately known whether this is linked to the trial drug or another unrelated factor, but the problems must be addressed and possible contraindications avoided, which may necessitate “de-blinding” (DuBois, 2008).

Although many  people would perhaps like to benefit from a new drug which is more effective than existing drugs, people have different ideas about what is an acceptable risk and different reasons for taking part in clinical trials. People who receive the placebo are not exposed to the same potential risks as those given the experimental drug. On the other hand, they have no possibility to benefit from the advantages the drug may offer. Those receiving a drug commonly considered as the standard therapy are not necessarily better off than those receiving a placebo as some participants may already know that they do not respond well to the accepted treatment (DuBois, 2008).

If people who participate in a clinical trial are not informed which arm of the trial they were in, valuable information is lost which might have otherwise contributed towards to treatment decisions made after the clinical trial. Taylor and Wainwright (2005) suggest that “unblinding” should occur at the end of all studies and so as not to interfere with the analysis of data, this could be done by a person who is totally independent of the analysis. This would, however, have implications for open label extended trials as in that case participants, whilst better equipped to give informed consent would have more information than the researchers and this might be conveyed to researchers in anad hocmanner.

Open label extension trails

Open label extension studies (mentioned in sub-section 7.1.8) seem quite fair as they give each participant the opportunity to freely consent to continuing with the study in the full knowledge that s/he will receive the experimental drug. However, Taylor and Wainwright (2005) have highlighted a couple of ethical concerns linked to the consent process, the scientific value of such studies and issues linked to access to drugs at the end of the prior study.

With regard to consent, they argue that people may have had a positive or negative experience of the trial but do not know whether this was due to the experimental drug, another drug or a placebo. They may nevertheless base their decision whether to continue on their experience so far. For those who were not taking the experimental drug, their experience in the follow-up trial may turn out to be very different. Also, if they are told about the possibility of the open label extension trial when deciding whether or not to take part in the initial trial (i.e. with the implication that whatever group they are ascribed to, in the follow-up study they will be guaranteed the experimental drug), this might induce them to participate in the initial study which could be considered as a form of subtle coercion. Finally, researchers may be under pressure to recruit as they can only recruit people in an open label extended trial who took part in the initial study. This may lead them in turn to put pressure (even inadvertently) on participants to continue with the study. 

The scientific validity of open label extension trials is questioned by Taylor and Wainwright (2005) on the grounds that people from the experimental arm of the first study who did not tolerate the drug would be unlikely to participate in the extension trial and this would lead to bias in the results. In addition, open-label trials often lack a precise duration other than “until the drug is licensed” which casts doubt on there being a valid research purpose.

The above authors suggest that open label extension studies are dressed up marketing activities which lack the ethical justification for biomedical research which is the prospect of finding new ways of benefiting people’s health. However, it could be argued that the aim of assessing long-term tolerability of a new drug is a worthwhile pursuit and if conducted in a scientific manner could be considered as research. Moreover, not all open label extension trials are open-ended with regard to their duration. The main problem in interpreting open label extension studies is that little is known about the natural course of the disease.

Protecting participants’ well-being at the end of the clinical trial

Some people who participate in a clinical trial and who receive the experimental drug experience an improvement in their condition. This is to be hoped even if benefit to the health of individuals is not the aim of the study. However, at the end of the study, the drug is not yet licenced and there is no legal right to continue taking it. This could be psychologically disturbing to the participants in the trial and also to their families who may have seen a marked improvement in their condition.

Taylor and Wainwright (2005) suggest that the open label trials may serve the purpose of prescribing an unlicensed drug on compassionate grounds, which whilst laudable, should not be camouflaged as scientific research. Rather governments should take responsibility and set up the appropriate legal mechanisms to make it possible for participants whose medical condition merits prolonged treatment with the experimental drug to have access to it.

Minimising pain and discomfort

Certain procedures to which people with dementia or their representatives consent may by burdensome or painful or simply worrying but in accordance with the principles of autonomy or justice/equity, people with dementia have the right to participate. The fact that they have made an informed decision to participate and are willing to tolerate such pain or burden does not release researchers from the obligation to try to minimise it. For example, if repeated blood samples are going to be necessary, an indwelling catheter could be inserted under local anaesthetic to make it easier or medical staff should provide reassurance about the use of various scanning equipment which might be worrying or enable the person’s carer to be present. In order to minimize fear, trained personnel are needed who have experience dealing with people with dementia. The advice of the carer, if there is one, could also be sought.

Drug trials in countries with less developed safeguards

Clinical trials are sometimes carried out in countries where safeguards are not well developed and where the participants and even the general population are likely to have less possibility to benefit from the results of successful trials. For example, some countries have not signed the Convention for the Protection of Human Rights and Dignity of the Human Being with regard to the Application of Biology and Medicine (1997) (referred to in section The participants in those countries may be exposed to possible risks but have little chance of future medical benefit if the trial is successful. Yet people in countries with stricter safeguards for participants (which are often richer countries) stand to benefit from their efforts and from the risks they take, as they are more likely to be able to afford the drugs once developed. This raises ethical issues linked to voluntariness because there may be, in addition to the less developed safeguards, factors which make participation in such trials more attractive to potential participants. Such practices also represent a lack of equity in the distribution of risk, burden and possible benefit within society and could be interpreted as using people as a means to an end.

Parallels can also be drawn to the situation whereby people in countries where stem cell research is banned profit from the results of studies carried out in countries where it is permitted or to the results of studies carried out in countries where research ethics are slack or inexistent.

For a detailed discussion of the ethical issues linked to the involvement in research of people in other countries, particularly lower and middle income countries where standards of protection may by lower, please refer to the afore-mentioned report by the Presidential Commission for the Study of Bioethical Issues.

[1] Please see

[2] Please see

  • Researchers should consider including a placebo arm in clinical trials when there are compelling and sound methodological reasons for doing so.
  • Researchers should ensure that patients are aware that the aim of a randomised controlled trial is to test a hypothesis and provide generalizable knowledge leading to the development of a medical drug or procedure. They should explain how this differs from medical treatment and care which are aimed at enhancing the health and wellbeing of individual patients and where there is a reasonable expectation that this will be successful.
  • Researchers should ensure that potential participants understand that they may be allocated to the placebo group.
  • It should not be presumed that the treating doctor or contact person having proposed the participant for a trial has been successful in communicating the above information.
  • Researchers conducting clinical trials may need training in how to ensure effective communication with people with dementia.
  • Appropriate measures should be taken by researchers to minimize fear, pain and discomfort of participants.
  • All participants should, when possible, preferably have the option of receiving the experimental drug (if proven safe) after completion of the study.
  • Pharmaceutical companies should not be discouraged from carrying out open-label extension studies but this should not be the sole possibility for participants to access the trial drug after the end of the study if it is proving beneficial to them.
  • In multi-centre clinical trials, where data is transferred to another country in which data protection laws are perhaps less severe, the data should be treated as stated in the consent form signed by the participant.



Last Updated: jeudi 29 mars 2012


  • Acknowledgements

    The above information was published in the 2011 Report "The ethics of dementia research" as part of Alzheimer Europe's 2011 Work Plan which received funding from the European Union in the framework of the Health Programme.
  • European Union


  • Send this page to a friend