Ethics Forum. Aug. 5, 2002.
Scenario: Does a doctor-patient relationship arise from online moderating?
We asked a physician who is an online moderator to a patient support group on the Internet to comment on the professional obligation to intervene when a posting suggests pending or immediate harm to the patient.
Reply:
As the public increasingly turns to the Internet for information and support, clinicians increasingly have the opportunity to incorporate the Internet into their work.
This is a new role for clinicians, and the obligations that attach may be unclear for them as well as the public. A significant concern with an Internet support group that is moderated by a physician is that individuals may expect and therefore wait for the moderator to intervene. If the moderator believes he or she is under no obligation to intervene and does not act, there could be an adverse health outcome and the moderator could conceivably be held liable for damages.
Moderating a group generally means maintaining the supportiveness of the group milieu. A moderator also may take on other roles, from which other obligations could flow.
As highlighted in an earlier Ethics Forum column (See column, Aug. 6, 2001), a key question is whether moderating an Internet support group creates a clinician-patient relationship. Such a relationship may exist if (a) the moderator offers and a group member requests diagnostic, prognostic, or therapeutic services or (b) even if not explicitly offered, the moderator actually does provide such services. Providing general information does not by itself create a clinician-patient relationship.
If the moderator only "moderates," i.e., facilitates discussion without contributing substantively to it, a clinician-patient relationship does not exist, and the moderator has no professional obligation to intervene. If, however, a clinician-patient relationship is created, the moderator has the same responsibilities he or she would have if the member were his or her office patient, including the professional obligation to intervene.
Psycho-Babble (http://www.dr-bob.org/babble/) is the online peer support group that I moderate. It is a true peer support group in that the support comes from members and not from me. As the moderator, my responsibility is to maintain the supportive milieu. If the milieu is at risk (for example, members become hostile), I am obligated to intervene, but if a member is at risk, it is the role of other members to provide support.
The moderator should define his or her role clearly and then should act according to that role. If the group is described as offering support from peers, but in fact the moderator gives medical or psychological advice, a group member may reasonably expect the clinician to intervene and the clinician may be so obligated.
When moderating Psycho-Babble, I make it clear in the Frequently Asked Questions and during the required registration process that I only moderate, and in fact I do not provide psychological or medical advice. Also, each main page links to a list of online crisis resources and each main page and the FAQ include the reminder, "In an emergency, please also get help in person."
Even if not professionally obligated to intervene, the moderator may choose to do so. In an emergency, Good Samaritan statutes may protect the moderator from liability. It is possible, however, that a pattern of intervening in emergencies may give rise to a reasonable expectation of intervention and therefore an obligation.
It should be noted that the ability of the moderator to intervene online might be significantly limited. Online, the moderator can be supportive and provide information and advice, but evaluation may be more difficult. The moderator and the group member may not be online at the same time, and of course the moderator may not know the member's name or location.
A clinician moderator's professional obligation to intervene depends on whether a clinician-patient relationship exists. If a group member receives (or could reasonably expect to receive) diagnostic, prognostic or therapeutic services from the moderator, the moderator has all of the obligations of a treating clinician, including that of intervening. If the only role is to moderate the group, the clinician is not obligated to intervene.
--Robert C. Hsiung, MD
Associate Professor of Clinical Psychiatry, University of Chicago
Back to top
Scenario: Is it ethical to give patients a placebo for pain treatment?
Some clinicians rely on placebos as a therapeutic tool to assess and try to relieve their patients' pain. What are the ethical implications of this practice?
Reply:
Concern over placebo use in clinical care is both apt and timely since it appears that this practice is ongoing, although data on its prevalence are not available.
When considering the use of placebos for "therapeutic" purposes, circumstances must be identified that ethically could justify the use of an agent that is without a predictable physiological or psychological effect, given that the usual intention of the therapeutic intervention in pain management is to provide a specific outcome (e.g., curative, palliative or rehabilitative).
Generally speaking, there are two types of clinical scenarios where placebos are used outside of formal investigational protocols: one legitimate (if done properly); the other inappropriate and possibly dangerous.
In the first instance, a placebo is used for diagnostic purposes in a manner similar to an "n of 1" trial. The patient serves as his or her own control to distinguish specific pharmacological effects from treatment-mediated suggestibility, prior to a recommendation for a potentially harmful, risky or costly procedure.
An example of this is the introduction of a placebo (e.g., saline) in a graduated diagnostic spinal anesthetic assessment of a patient with chronic pelvic pain without determinable pathology, before recommending surgical exploration or other interventional therapies.
Ethical considerations in these rare instances parallel more formal research studies. Disclosure and consent that an agent with no specific pharmacological activity will be introduced as part of the procedure and testing is obligatory. This approach provides the ethical foundation for proceeding, even though interpretation of results from these types of procedures may be confounded in highly suggestible patients.
Of far greater concern, and apparently a more common practice, is the use of placebos as a "therapeutic" alternative or simply to fool the patient into thinking he or she is receiving active therapy. The most egregious example I recently encountered was the use of IM saline injection in postoperative abdominal and thoracic surgery patients, because a particular group of surgeons feared that they would "turn their patients into addicts" if they used opioid analgesics.
The reasons behind physicians' decisions to use placebos in this manner are varied, but they generally reflect a poor understanding of placebo mechanisms and constitute serious ethical breaches. Specifically, the ethical concerns fall into two broad categories: absence of informed consent and risk of creating harm. An individual's right to know what is being done and to participate in decision-making through informed consent is at the center of contemporary medical ethics and is also reflected in the law.
Potential harm can range from risks of physiologic morbidity from inefficacious treatment, to discouragement and demoralization from "treatment" failure, to irreparable loss of trust in the fiduciary patient-physician relationship.
If patients learn that they have been fooled intentionally, how will they be able to regain confidence in the medical profession? Although difficult to measure, this betrayal may carry more profound and enduring harm, negatively impacting the current and future relationships between the patient and health care professionals, including the willingness to seek help when it is needed.
It is true that placebo-induced analgesic responses can be powerful, exactly replicating potent pharmacological effects under some circumstances. But these responses are highly dependent on the conditioning environment, the suggestibility of the patient and a host of other still-to-be determined factors.
It is interesting to note that there seems to be an absence of appreciation for the potential of pain to cause morbidity compared with other pathophysiological states. For instance, it would be viewed as untenable to use placebos to "treat" hypertension, diabetes or asthma, as well as to use placebos to determine "how real" the symptoms associated with these disorders are. As in these other conditions, placebo effects in pain treatment are too variable, inconsistent and unpredictable, and the potential physiological and psychological harm from poorly managed pain too great to rely on them in lieu of effective pain-relieving therapies.
There are important and justifiable uses for placebos in controlled clinical trials involving pain assessment or pain relief when two crucial conditions have been met: informed consent has been obtained and the risks of harm or suffering are predictably negligible. In clinical settings, however, the practitioner's role is not specifically to advance science, but rather to provide care in the patient's best interest. The physician would need to confront not only the magnitude of medical harm likely to be caused by use of a placebo, but also, as discussed, the potential harm from deception.
The relief of pain and other distressing symptoms is a fundamental duty of physicians. Use of placebos in clinical practices marginalizes patients with pain, disregarding it as neither a highly consequential symptom nor a disease state of the peripheral and/or central nervous system as it is now known to be. Failure to use effective therapies in favor of placebos puts patients at risk and makes the doctor highly vulnerable, subject to both ethical and legal sanctions.
--Perry G. Fine, MD
Professor of Anesthesiology, Associate Medical Director, Pain Management Center, University of Utah School of Medicine, Salt Lake City, National Medical Director, VistaCare
Ethics Forum discusses questions on ethics and professionalism in medical practice. Readers are encouraged to submit questions and comments to philip.perry@ama-assn.org or to Ethics Group, AMA, 515 N. State St., Chicago, IL 60654; fax 312-464-4613. Opinions in Ethics Forum reflect the view of the author and do not constitute official policy of the AMA.
Back to top
Copyright 2002 American Medical Association. All rights reserved.
SHARE
