Feeding back the findings

People who volunteer to take part in research projects make significant and altruistic contributions. The projects will obtain information that may be of interest and importance to these participants. Sometimes it will have been made clear to them, before they gave their prior consent, that they will be informed of the findings. The examples below are taken from medical research, but the value of the ensuing discussion is transferable to other fields of research.

Clinical research

Most participants in clinical research studies are patients, and in most cases they will be informed of their treatment and clinical status in the same way as in the normal health service. However, there are exceptions when it comes to clinical trials. These trials often split participants into two groups through so-called randomisation (random allocation into groups). One receives the standard treatment (or a placebo treatment) while the other group receives the new treatment which is being tested. In order to avoid a certain form of systematic error, information bias, such trials are often double-blind. This means that neither those who measure the subjects' clinical status during the course of the trial nor the participants themselves know which treatment the individual subject is receiving.

The ethical dilemma is associated with deciding when to stop the trial and inform the subjects of which group they were in. If the new treatment is better than the current one, it is important that both the placebo group and everyone who suffers from the same disease receives the new treatment. However, for scientific reasons it may be important for the trial to continue long enough to allow the researchers to calculate the efficacy of the new treatment to a high level of precision and until it has been determined whether it has serious, but rare, side effects. The decision to discontinue a trial is often made on the advice of an external committee.

Epidemiological research

This type of study is based on a sample of the population and the subjects are commonly in good health.

Trials are also run with subjects who are generally healthy, such as vaccination trials. A randomised controlled trial of a vaccine against meningococcal disease was conducted with 14 and 15-year-old participants in Norway in the late 1980s. Once the results had been analysed and a certain positive effect had been established, the trial's placebo subjects were offered the active vaccine, which many accepted. This meant that the subjects learnt which group they had been in. This was a positive move with respect to personal protection against disease, but it ruined the trial design and later follow-ups. When suspicion arose many years later that Myalgic Encephalomyelitis/Chronic Fatigue Syndrome (ME/CFS) might be a side effect of the vaccine, the situation was difficult to untangle. It can be considered an ethical problem that many trials are discontinued too early or that their clinical endpoints are not very relevant. This introduces uncertainty with respect to whether the treatment is actually efficacious or whether it has serious side effects. However, this aspect must also be balanced against the benefit to participants (and everyone else in the same risk situation) of learning which group they were in and receiving the best possible treatment.

Other designs include cross-sectional studies, case-control studies and cohort studies as well as more complicated designs. If the observations are in the form of questionnaires completed by the subjects, there is little point in feeding back information. The situation is different if blood samples have been obtained or data have been collected from other sources. These situations may give rise to new information that may be of interest to the participants and which sometimes may be highly significant. As a general principle it is difficult to envisage situations that warrant non-disclosure of such information to the participants. This disclosure principle means that the subjects have right of access to the information, but not necessarily that researchers are duty-bound to provide all the details. Another good principle is to routinely produce newsletters, and to ensure that research results are presented in a simple and understandable way.

Researchers are keen to work with large samples. One way to achieve this, is to tempt people with useful medical checks for participants, such as taking blood pressure or cholesterol readings, or perhaps even more sophisticated examinations which participants will not have access to through the ordinary health service. In such cases, all participants must receive detailed information about the benefits and drawbacks of such tests. They should be informed that the test might reveal a high risk of serious disease. This may be important if there is a treatment available that can prevent this serious disease. If there is no such treatment, the individuals concerned will receive information that may cause considerable fear and anxiety. If the information includes knowledge about genes that increase the risk of disease, it may also have implications for close relatives. People respond in a variety of ways to learning about the risk of disease, and these reactions cannot always be foreseen before the results are on the table. In general, researchers should take care to avoid using new information as bait to boost participation, just as no financial benefit should ensue from taking part in a research project.

Ideally, participants should be allowed to decide in advance whether or not they want to receive feedback. Each individual will then be in a position to decide whether he or she wishes to donate data and biological material to research, and to decline the offer of receiving information about themselves. Others may be highly motivated to receive information about themselves in return for their contribution.

What is the best course of action if an agreement has been made with a participant that no findings should be fed back, but the researcher holds information which is considered important for the participant to obtain? It is difficult to give general advice. The specific situation will always need to be assessed and a number of considerations taken into account. This assessment should be made by the ethical research committee that originally vetted the project.

In certain situations insurance companies, employers or police officers may want access to information generated by a research project. Researchers should comply with a code of conduct that precludes their disclosing such information to anyone but the participants themselves. A dilemma arises when participants are pressurised by others. Only the subjects themselves have right of access to the data. It is worth noting that under the Biotechnology Act nobody but the person concerned has a right to request or use genetic information.

Imagine a situation in which an accident has caused a number of people to go missing. Biological material is available from unidentified casualties, as in the tsunami disaster. In this context one solution might be to link the list of missing persons to the register of research subjects who have taken part in projects that collect and store DNA. Prior to making decisions of this nature, it is important to obtain an ethical assessment from sources external to the project management, first and foremost from the research ethics committee that approved the project.

This article has been translated from Norwegian by Hege Hernæs, Akasie språktjenester AS.