Establishing guidelines for Internet-based prescribing. (For Debate).

Abstract: The American Medical Association called for the establishment of guidelines to allow electronic prescription of medications "for established patients." Based on experience writing more than 10,000 Internet-based prescriptions, we agree that guidelines are long overdue. Restricting such prescribing to patients with whom a face-to-face relationship has previously been established violates patient autonomy and distorts the physician-patient relationship without improving patient safety or convenience. A study comparing information obtained and used by Internet-based physicians prescribing sildenafil with that obtained by clinic-based physicians writing similar prescriptions suggests that safety may be greater on the Internet. Data regarding the appropriateness of prescriptions for other medications suggest that the in-office visit is not the panacea it is often assumed to be. Guidelines for electronic prescribing, like guidelines for other aspects of medical practice, need to be based on evidence. Such evidence is currently lacking, and a serious effort to obtain it should be a top priority.

Editor's Note: Occasionally we receive articles that address controversial issues. We will publish some of these articles under the heading "For Debate." We hope that the publication of these articles will generate discussion among readers and stimulate the submission of letters to the editor. This article is an example. We emphasize that the Southern Medical Journal's editor-in-chief and its editorial staff, as well as the Southern Medical Association, neither condone nor condemn the content of this article. We are publishing it because it pertains to a relevant issue in the practice of medicine, and we trust that many of our readers will find it of interest. We look forward to receiving comments about it from our readers.

At its most recent meeting, the American Medical Association's House of Delegates voted, "That the AMA develop guidelines to assist physicians in using the Internet for legitimate electronic prescribing of pharmaceuticals for established patients." (1) Guidelines are in fact long overdue. Much of the controversy that has swirled around online prescribing in recent years rests on the absence of guidelines and the consequent uncertainty about what standards are being applied. More than 10,000 Internet-based prescriptions for sildenafil written by one of us (MJJ) over the past 3 years indicates, however, that restricting electronic prescriptions to patients with whom a previous off-line relationship has been established would curtail the cherished principle of patient autonomy, prevent establishment of new types of patient-physician relationships, and severely limit the economic and efficiency benefits of Internet prescribing-all without improving patient safety or convenience.

Respect for patient autonomy and the patient-physician relationship are fundamental principles of medical ethics. (2) Autonomy includes the patient's right to choose the physicians from whom he receives care. The patient-physician relationship establishes the framework of trust and responsibility under which the care will be rendered. The same House of Delegates meeting that recommended development of guidelines for online prescribing adopted a report on the patient-physician relationship (3) that, after quoting a medical ethicist on the variety of such relationships "given the complexity of professional styles, patient expectations and values, and contexts in which the relationship is established," (4) goes on to say, "Irrespective of the circumstances of the encounter between patient and physician, medical ethicists have characterized it as a moral activity." More specifically, the part of the report officially adopted as part of the Code of Ethics states, "The relationship between patient and physician is based on trust," and "A patient-physician relationship is generally created by mutual agreement between physician and patient." Taken together, these various statements reveal no basis for attempting to prevent patients from seeking online medical care from physicians of their choice, regardless of previous relationship. The essence of the patient-physician relationship is trust and agreement, not a face-to-face encounter, and a reasoned medical decision is based on sufficient information, not on how that information is obtained.

Why Patients Seek On-line Consultation

Sildenafil provides a particularly cogent example of why patients may seek consultation from someone other than their usual physician. Many men have been led to consider the firmness of their erections a sign of their manhood. Others may have been brought up to believe that one does not discuss such matters with someone of the opposite sex--even if that someone is a physician. For still others, it is simply too private and too intimate. For all of these reasons and others, patients may prefer to consult with a physician who does not personally know them.

Yet, face-to-face alternatives are not always easy to find. In rural areas, there may be no physician other than the one they usually see--often a personal friend. Individuals in military service, especially if stationed overseas, may also find it difficult to arrange an appointment with an appropriate nonmilitary physician. Patients may have special reasons for wanting the added psychologic and social distance of Internet-based prescribing. One patient of the NET Doctor group, with which one of us (MJJ) is associated, is a professional-level employee at a large Swiss pharmaceutical company. One may speculate that he' feels particularly compelled to ensure that his colleagues never hear he has sought a prescription for another company's product, and finds it reassuring that the physician he consulted will never encounter his colleagues in person.

Other individuals may prefer an Internet-based physician for the same reason the Roman Catholic Church places a screen in the confessional--the absence of face-to-face contact makes the encounter less personal and encourages openness. One patient said, in so many words, "I can't bring myself to see a doctor." However, since he did not have to literally see the doctor, he was able to talk about his problem in considerable detail. Many other patients have also been open and candid about details of their sex life and about how their erectile difficulties are affecting their relationship with their partners. Any physician in a physical office setting who encountered the same degree of candor would be justified in concluding that he had fully won the patient's trust.

Another category is patients who have consulted other physicians, but with less than satisfactory results. As the editor of Internet Medicine remarked regarding a recently published study of Internet sildenafil prescribing, (5) "if a primary caregiver is unwilling to discuss intimate matters of importance to patients, or prescribes treatments that are slow and expensive (like counseling) or simply don't work, patients turn to the Internet." (6)

One such patient noted how, over the previous 2 years, he had less and less frequently had firmness enough for sex. "Doctors say I am fine," he wrote, "but my wife wants, as I do, more." Unlike the physicians he had previously consulted, NET Doctor Group took his problem seriously.

There are also the patients, especially younger men, who have no current physician and see no need for one. With a single condition requiring medical attention, they prefer to go to a physician on an as-needed basis when it is convenient. Perhaps that is why the NET Doctor Group has had patients in Rochester, MN, who could choose from among the country's highest per capita physician population. Conversely, one of the authors (MJJ) has been consulted by a patient who just a month earlier had had a kidney transplant. He obviously trusted his transplant physicians but did not want to discuss his sex life with them. (As it happens, the Internet questionnaire and follow-up telephone interview did not provide sufficient information to justify a sildenafil prescription so soon after the operation. This may be one instance in which an in-office visit, with opportunity for a thorough physical examination, would have allowed the prescription to be written.)

Often, patients come to NET Doctor Group (as they might visit a physician in a physical office) who have had previous sildenafil prescriptions. Some patients may have been dissatisfied with their previous physician's services--that may have been why one such patient said he did not want to "renew locally." Other patients say they have just moved to a new locale or have had their previous relationship disrupted by a change in insurance coverage and have not yet found a new physician.

There may be other reasons as well. In a particularly interesting case from the literature, (7) a man's wife confiscated the supply of sildenafil that his regular physician had prescribed, intending to dole it out as she saw fit. Desiring a supply under his personal control, and not wishing to explain all this to his regular physician, he sought (and obtained) a telephone prescription from a physician acquaintance he did not see as a patient.

The NET Doctor Group gets patients who say they are seeking "less embarrassing" ways to purchase sildenafil or are looking for a lower cost. These expressions suggest a visit to their physician's office was not satisfactory or physician fees for renewal are significantly higher than those found on the Internet. This raises the question of whether such patients have a right to demand that their contacts with one physician not be reported to another. Today, they assume this from the nature of the online consultation. Tomorrow it may be a matter of explicit patient choice.

In an ideal world, every patient would have a physician he trusted to manage all aspects of his medical care, even the most intimate. Unfortunately, we do not live in an ideal world. For reasons that may have nothing to do with an individual physician's competence or empathy, some patients may choose to go elsewhere for certain aspects of their care. The principle of patient autonomy dictates that their wishes be respected.

Safe and Appropriate Prescribing

Since patients demand Internet-based prescribing, guidelines must focus on identifying the information needed to safely and appropriately prescribe a given medication. This will obviously vary with the medication. There are some medications, such as the benzodiazepines and narcotics, that have a high potential for abuse. Such medications can appropriately be prescribed only in a setting where risk of abuse and misuse (which frequently occur even in the office-based setting) is significantly reduced.

Most medications are intended to treat specific diagnostic etiologies. In this case, the initial question becomes: What information is required to make the diagnosis and successfully treat the patient? In most instances, the diagnosis will require physical examination or laboratory tests or both. Clearly, such diagnoses cannot be made or treatment initiated without physical contact with the patient. In some instances, diagnoses are routinely made on the telephone. This usually happens when symptomatic reports point clearly to a common disorder that can be treated on an empirical basis. Or perhaps, as with louse infestations, patient descriptions may be sufficient to establish a definite etiology. Clearly, such disorders can equally well be diagnosed on the Internet, as the AMA recognized in its report.

In other cases, medications are intended to treat symptoms rather than causes. Antihistamines are one example; sildenafil is another. In such cases, it is only necessary to establish that the symptom exists and that there are no contraindications.

The key to establishing symptoms and detecting contraindications is a careful medical history. Using the Internet, the history is typically taken through a carefully structured questionnaire that may run to more than 50 detailed items. In an office visit, by contrast, the history-taking is likely to be largely free-form. Research has shown that structured questions elicit far more complete and detailed information. For example, Locke et al (8) found that substituting a computer-administered questionnaire for a standard American Red Cross interview increased the proportion of donors identified as having risk factors for HIV transmission from 0.13% to 4.4%. Owens and Dalzell (9) found that in a general pediatric clinic, asking questions about five specific sleep behaviors rather than a single open-ended question more than tripled the proportion of children identified as having bedtime issues and also significantly increased the proportion identified as snoring or exhibiting nighttime awakenings. Structured ques tionnaires have also been shown to improve identification of psychiatric disorders (10) and of alcoholism. (11)

There may be occasions when a patient seeks to have a prescription renewed by a physician other than the one who initially issued it. So long as the patient provides sufficient information to allow the physician to make a reasonable medical decision, the appropriateness of the renewal should be clear.

The question is sometimes raised whether patients cannot lie more readily on the Internet. Physicians typically assume patients are telling the truth to the best of their ability. Only inconsistent responses elicit follow-up questions, just as inconsistent responses to an Internet questionnaire elicit a follow-up telephone interview. In contrast to the typical office setting, the NET Doctor Group specifically warns online patients that any false or incomplete answer may have adverse or even fatal consequences.

In summary, the key to appropriate prescribing is obtaining sufficient information to allow the physician to make a reasoned medical decision, not how that information is obtained.

Office Visits: No Panacea

Gunther Eysenbach, (12) editor of the Journal of Medical Internet Research, recently wrote the following:

FDA evidence for the alleged risks of online prescribing to date merely consists of a few anecdotal cases. The most frequently cited case is the story of a 52-year-old Illinois man with episodes of chest pain and a family history of heart disease, who died of a heart attack in March 1999 after buying Viagm (sildenafil citrate) from an online source that required only a completed questionnaire to qualify for the prescription. Though there is no proof linking the man's death to the drug, FDA officials say that a traditional doctor-patient relationship, along with a physical examination, may have uncovered any health problems such as heart disease and could have ensured that proper treatments were prescribed. However, it should be noted that there have been several similar cases where patients with a comparable history have died while taking Viagra, despite receiving their prescriptions at the doctor's office.

What is theoretically possible in a face-to-face visit is often not reflected in the real-life behavior of busy physicians, as is confirmed by a study published in the same issue of the journal just cited. (5) This study contrasted the information available on the first 2,104 patients seeking sildenafil prescriptions from the NET Doctor Group with that in the medical records of the 36 patients who received sildenafil prescriptions during the same period at the clinics of an innercity teaching hospital. Of these 36 records, only 20 recorded a general physical examination during the previous 6 months (including the prescribing visit) and lust 16 included a complete list of medications being taken. Blood lipid profiles and tests for diabetes were uncommon, and descriptions of the patient's sex life and erectile function tended to be sketchy if present at all. It is thus obvious that few clinic physicians took advantage of the theoretical opportunity to seek out the causes of the erectile dysfunction or to counse l the patient, and many were remiss in determining or documenting whether medications that could constitute a contraindication to sildenafil use were being taken.

The conclusion that office visits are no panacea is reinforced by data on inappropriate prescription of other medications. For example, in-home interviews of community-dwelling elders found that between 22.5% and 29.0% were taking at least one inappropriate medication. (13) Tamblyn et al (14) used a "standard patient" approach to examine prescribing of nonsteroidal anti-inflammatory drugs (NSAIDs), finding that 41.7% of the visits by a patient with symptoms of early osteoarthritis resulted in inappropriate prescriptions. The risk of an inappropriate prescription was significantly increased when contraindications to NSAID therapy were incompletely assessed. Using retrospective chart review, Pickering et al (15) found that only 25% of the prescriptions for ciprofloxacin in an academically oriented long-term care facility were appropriate. Twenty-three percent of the prescriptions were written in the absence of any indication for this medication, while in 49% of the cases an alternative medication was cheaper or more effective or both.

Recent data suggest that the office visit is not effective in ensuring accurate diagnosis and treatment of allergies; 65% of the patients given antihistamine prescriptions for presumed allergies had no IgE reaction to common allergens. (16) Batty et al, (17) applying an indicator algorithm to judge the appropriateness of benzodiazepine prescriptions for elderly patients at 17 English and Welsh hospitals, found that 65% of the prescriptions examined were inappropriate.

Clearly, ways to help physicians prescribe more appropriately must be sought. However, it is equally clear that it is inappropriate to judge Internet-based prescribing by an ideal of in-office prescribing that is rarely approached in practice. Focused Internet prescribing may be one way of achieving a higher standard of practice.

Need for Further Research

Guidelines for online prescribing, like clinical guidelines, should be based on solid evidence. Unfortunately, such evidence is largely lacking. The study by Jones (5) is the only one to report an extensive patient series. Even this study was limited by the fact that patients, having been promised they would receive no unsolicited email, could not be recontacted for follow-up.

Papers published in the New England Journal of Medicine (18) and Annals of Internal Medicine (19) cast a negative light on Internet prescribing but provide no direct evidence of patient harm or adverse outcome. These reports are not based on clinical experience, and they use an unverified clinical standard that could almost be called a dogma. As Bysenbach said in his editorial:

This is the bottom line: Currently, we simply do not have sufficient evidence whether, and under which conditions, online prescribing of relatively safe drugs such as the impotence drug Viagra (sildenafil citrate) actually creates more harm than benefit, or vice versa. More research is urgently needed to address questions such as which drugs can be prescribed safely and to which kinds of patients, and which safeguards we can install to monitor adverse events.

Eysenbach concluded his editorial by calling for more research. The only conclusion of this editorial is to echo his call and to encourage investigation into the relationships among prescribing paradigms, practice reality, and the Internet.

Accepted January 25, 2002.

References

(1.) Report 4 of the Council on Medical Service. Medical care online. Presented at the annual meeting of the American Medical Association House of Delegates, Chicago, June 17-21, 2001.

(2.) American Medical Association, Council on Ethical and Judicial Affairs. Opinion 10.01: Fundamental elements of the patient-physician relationship. Code of Medical Ethics: Current Opinions with Annotations, 2000-2001. Chicago, AMA Press, 2000.

(3.) Report of the Council on Ethical and Judicial Affairs. The patient-physician relationship. Presented at the annual meeting of the American Medical Association House of Delegates, Chicago, June 17-21, 2001.

(4.) Thomasma DC. Beyond medical paternalism and patient autonomy: A model of physician conscience for the physician-patient relationship. Ann Intern Med 1983; 98:243-248.

(5.) Jones MJ. Internet-based prescription of sildenafil: A 2104-patient series. J Med Internet Res 2001;3(l):E2.

(6.) Ward BO. Even without peer review, Viagra research shows doctors why patients become Internet-positive. Internet Med 2000;5(6): 1,7.

(7.) Finkbeiner A. Random musing about a little blue pill. J Ark Med Soc 1998;95:l00-l0l (editorial).

(8.) Locke SE, Kowaloff HB, Hoff RG, et al. Computer-based interview for screening blood donors for risk of HIV transmission. JAMA 1992;268: 1301-1305.

(9.) Owens JA, Dalzell VP. The "BEARS" screening for pediatric sleep problems in the primary care setting. Sleep 2001;24:A216 (abstr).

(10.) Staab JP, Datto CJ, Weinrieb RM, et al. Detection and diagnosis of psychiatric disorders in primary medical care settings. Med Clin North Am 2001:85:579-596.

(11.) Perdrix A, Decrey H, Pecoud A, Bumand B, Yersin B. Detection of alcoholism in the medical office: applicability of the CAGE questionnaire by the practicing physician: Group of Medical Practitioners PMU [in French]. Schweiz Med Wochenschr 1995;125: 1772-1778.

(12.) Eysenbach G. Online prescriptions of pharmaceuticals: Where is the evidence for harm or for benefit? A call for papers--and for reflection. J Med Internet Res 2001;3(1):E1 (editorial).

(13.) Hanlon JT, Fillenbaum GG, Schmader KE, Kuchibhatla M, Homer RD. Inappropriate ding use among community-dwelling elderly. Pharmacotherapy 2000;20:575-582.

(14.) Tamblyn R, Berkson L, Dauphinee WD, et al. Unnecessary prescribing of NSAIDs and the management of NSAID-related gastropathy in medical practice. Ann Intern Med 1997;127:429-438.

(15.) Pickering TD, Gurwitz JH, Zaleznik D, Noonan JP, Avom J. The appropriatenses of oral fluoroquinolone-prescribing in the long-term care setting. J Am Geriatr Soc 1994;42:28-32.

(16.) Szeinbach S, Boye M, Muntendam P. O'Conner R. Diagnostic assessment and resource utilization in patients prescribed non-sedating antihistamines. Presented at the annual meeting of the American College of Osteopathic Family Physicians, Philadelphia, March 28, 2001.

(17.) Batty GM, Oborne CA, Swift CG, Jackson SH. Development of an indicator to identify inappropriate use of benzodiazepines in elderly medical in-patients. Int J Geriatr Psychiatry 2000;15:892-896.

(18.) Armstrong K, Schwartz JS, Asch DA. Direct sale of sildenafil (Viagra) to consumers over the Internet, N Engl J Med 1999;341:1389-1392.

(19.) Henney JE, Shuren JE, Nightingale SL, McGinnis TJ. Internet purchase of prescription drugs: buyer beware. Ann Intern Med 1999; 131:861-862 (editorial).

RELATED ARTICLE: Key Points

* A valid patient-physician relationship may be established solely on the basis of electronic contact.

* The value and necessity of physical examination for all patient and all medical conditions is unsubstantiated.

* Internet prescribing is an extremely convenient, cost-effective, ostensibly safe system of medical practice suitable for some medications and patients.

* Non-Internet-based practice does not meet the needs of all patients; conversely, Internet prescribing is satisfactory, even preferable, for a substantial number of patients.

* Significant research on the relationship of the Internet to patient safety and prescribing must be conducted before practice guidelines and regulations for Internet prescribing are developed.

From Consultative & Diagnostic Pathology, Inc., Lee's Summit, MO, and Thomasson Editorial Services, Oak Park, IL.

Reprint requests to Miles J. Jones, MD, Consultative & Diagnostic Pathology, Inc., 1704 SE. 11th Street, Lee's Summit, MO 64081.

Copyright [c] 2003 by The Southern Medical Association