is not autonomous or self-determining.1 Murphy argues that this model which she calls the advocate model of the nurse-patient relationship, reflects the basic value of nursing which is the best possible care for patients.39 However, nurses should not be considered as being in a solely position to act as patients advocates. According to Johnstone all professions with a morally significant relationship with a patient ought to fulfil the role of the advocate.1 The ANA Code for nurses though, refers to the nursing role of advocate in safeguarding the well being of the client and the society.8 However, both the nature of nursing which focuses on caring, preventing harm and protecting dignity and the advocates role of nurses which calls for defending the rights of subjects, are sometimes incongruent with the ethics in research. Conflicts in nurses Beneficence-Non malificence A common feature in professional conduct codes and those specific to research is the principle of non-malificence. The ANA Code of conduct declares that the nurse protects the clients and the public from unethical, incompetent or illegal practice of any person.8 This statement raises the issue of advocacy when nurses have to protect patients from the researchers’ incompetence or unethical behaviour. Even if nurses are certain about the incompetence of the investigator, which is usually very difficult, they have to deal with serious dilemmas. First they have to consider the fact that if patient learn that they are exposed to professional misconduct, they may lose faith in health care. Jameton though, believes that patient should be informed as they will appreciate the trust shown to him by frankness.13 If the researcher does not inform or compensate patient then nurses have to decide between the duty to safeguard the well-being of patient and be loyal to them, and the loyalty to colleagues. However, even if nurses decide that their duty of caring and being loyal to the patient is more important, they may have to deal with the hierarchical and bureaucratic systems of institutions which demand loyalty to subordinates to the institution. In case the incompetent researcher is a higher status professional, nurses may be obliged to show loyalty, but this can conflict with loyalty to HEALTH SCIENCE JOURNAL ® VOLUME 5, ISSUE 1 (2011) What are the Major Ethical Issues in Conducting Research? Is there a Conflict between the Research Ethics and the Nature of Nursing? pp: 3-14 E-ISSN:1791-809X www.hsj.gr Health Science Journal® All Rights Reserved 10 patients. Consequently, nurses may feel that their patients are vulnerable and exposed and that they can not prevent it because they do not have a voice or power to resist. This is merely why many authors believe that it may not be possible for nurses to act as advocates of subjects in research. Many support the idea that the prohibition from the advocacy role comes from the origins and development of nursing as a women’s occupation dominated by medicine in a bureaucratic system.40 Another possible issue of conflict is that the caring nature of nursing with regard to the right of patients to the best treatment/care is sometimes conflicting with the aim of research in non therapeutic studies. According to the Belmond commission the general aim of practice is to enhance the well being of individuals while the purpose of research is to contribute to general knowledge. This distinction highlights the differences in the aims of a nurse practitioner and a researcher. It is therefore very difficult for nurses to be engaged in studies whose aim is not directly beneficial to the subject. They must though, consider that these studies may generate and refine nursing knowledge. Another problem that nurses may have to face is taking part in randomised control trials. According to Brink and Wood dedicated nurses are finding themselves under pressure when they are asked to exclude some patients from an obvious beneficial treatment such as relaxation techniques for relief of post operative pain.41 So, they suggest that whenever it is possible to predict such problems for nurses, the control data should be collected before introducing the beneficial variable. Skodol Wilson implies that there should