Oxford University Press's
Academic Insights for the Thinking World

Ethical issues in developing scales for biomedical research

When we think of ethical issues in biomedical research (if we think of them at all), what usually comes to mind are either egregious breaches, such as the infamous Tuskegee syphilis study, in which treatment was kept from rural black men in order to investigate the natural history of this treatable disease, or questions such as whether it is proper to use prisoners as subjects in early drug testing trials. It’s usually assumed that studies done for the purposes of developing and validating paper-and-pencil scales are immune from ethical concerns; after all, what problems could possibly arise in giving people a scale to complete? While it’s true that life and death issues are rarely present, there are in fact situations that can lead to ethical challenges. Let’s look at a few of them.

1. As part of the development of a scale of marital satisfaction, you administer the questionnaires to couples. Some of the items ask whether the person has ever had an extramarital affair. Needless to say, you assured participants that their responses would remain confidential. However, a few years later, one of the couples is going through a nasty divorce, and the lawyer for the husband has subpoenaed the wife’s questionnaire to use as evidence. Does the legal order trump your pledge of confidentiality?

2. You are trying to develop a new scale of suicidal ideation. The score for one adolescent is much higher than those of others in the sample. However, the scale has not yet been fully validated. Should you act on your concerns? If so, whom should you notify — the child, the parents, the family physician, someone else?

3. You are designing a test to measure motor coordination in children, which involves a number of tasks related to balance, ball-handling skills, and fine motor coordination, such as tracing with a pen and threading beads onto a string. The test is difficult for children with poor motor coordination and can cause anxiety, anger, and frustration. What should you do — not include such children in the development phase (which would jeopardize the norms), warn them ahead of time (which may affect their performance), counsel them afterwards, or switch to a different profession?

Researchers review documents. Public Domain via Wikimedia Commons
Researchers review documents. National Cancer Institute. Public Domain via Wikimedia Commons

4. Your aim is to develop a measure of self-esteem. In order to validate it, you design a study that involves assessing two groups of students before and after they make an oral presentation. Participants in the first group are told that they did very well, while those in the other group are told that they had done very badly and had made fools of themselves; this “feedback” was determined ahead of time and wasn’t dependent on their actual performance. Moreover, the students were enrolled in an Introductory Psychology class, and had to participate in three studies as part of the course requirements. Should this study be done? If so, what safeguards need to be in place? Are there concerns about requiring students to be participants in studies?

5. You have developed a new scale of parenting ability. It appears to show that those from a specific minority group have much poorer skills. Should you continue with the development of the questionnaire?

6. In order to develop a test of abstract reasoning ability, it is necessary to administer it to people ranging in age from five to 75 years, and to patients who may exhibit problems in this area, such as schizophrenics and those with brain injuries. What does it mean to get “informed consent” from people whose judgement may be compromised by age or psychiatric condition?

These situations raise a number of ethical issues — confidentiality, the duty to warn, the effects of the testing on the psychological well-being of the participants, the use of deception, the social consequences of the test results, and what “free and informed consent” means in actual practice. The perceptive reader will also note that we have been very good at raising questions about ethical concerns, but haven’t resolved any of them. In large measure, this is because there aren’t black-and-white answers, but only competing demands that must be balanced against one another and, in some areas, broad and evolving guidelines. For example, the first draft of Canada’s Tri-Council Policy Statement on Ethical Conduct for Research Involving Humans specifically forbade the use of deception in research. However, social scientists protested, stating that some valuable research would become impossible to conduct. As a consequence, the latest version does allow it, but mandates various procedures to protect the wellbeing of the participants. Also, there are guidelines that are usually followed when enrolling people who, because of intellectual or developmental problems, are unable to give consent for themselves (e.g., having a substitute decision maker). However, these often vary from one jurisdiction to another, and even among research ethics boards within the same state or province (not the most desirable of situations for multi-site projects).

The bottom line is that ethical issues do exist in the course of developing scales, and researchers need to be aware of the areas where problems may arise. When in doubt, talk to the ethics board of your institution before the study begins, so that they can be identified and hopefully prevented.

Header image: Filling out a survey for Rainbow Health Initiative by twodolla. CC BY 2.0 via Flickr.

Recent Comments

  1. Richard

    Does scientific research answers everything?? I mean people are doing it intentionally, it’s all in their head. Their actions speaks louder that words. Some got it in genes.

Comments are closed.