Psychiatrists cannot include everything they observe and everything the patient says in the medical chart. If they tried to include everything, the significant points would get buried in reams of irrelevant facts. Psychiatrists must choose what to include and what to omit. They must focus on certain problems, emphasize some points and neglect others, and phrase their entries in one way rather than another. In short, they must select and tailor what goes into the chart.
Psychiatrists should tailor the chart to focus on what is significant for the diagnosis and treatment of the patient. But sometimes psychiatrists tailor the chart for other purposes: to ensure that managed care will cover continued hospitalization, to secure a civil commitment, to obtain a long-term placement, or to protect themselves against malpractice claims. Is tailoring the chart for these purposes a form of deception? Is tailoring the chart for these purposes ever ethically justified? Our aim in this paper is to address these two questions.
Our interest in these questions began with some incidental remarks. Several medical students remarked to us that they were told during their clerkship in psychiatry to tailor charts for various purposes. They said that in some cases they were told not to note the patient's improvement in the progress notes. When we heard about this practice, we were concerned and curious. So we asked some students, residents, and attending psychiatrists at two medical centers about the practice of tailoring charts. A number of people told us that the practice did occur and gave us examples. We did not set out to determine the prevalence of the practice of tailoring charts. Our purpose in talking to various people was simply to get a sense of the practice and to gather some examples that we could analyze. Our purpose was and is to contribute to an ethical discussion, because we found little discussion in the psychiatric journals or in the two training programs we examined.
In this paper we discuss four cases that people told us about. In each case, an ethical problem exists because of competing "goods." On the one side is the good of honesty, forthrightness, and accuracy. On the other side is some other good to be achieved: a benefit for the patient, a benefit for others, fairness, or efficiency. The question to be discussed is whether the nature of the problem or the other good to be achieved justifies some sacrifice in honesty and accuracy.
If one believes that honesty is an absolute value—a concern that overrides all other concerns—then there is no need for a discussion. But most people don't believe that. They believe that honesty is an important value that may compete with other values. They can imagine cases in which dishonesty is ethically justified. For example, most people think it is ethically justified to lie to save the life of someone who is being hunted and persecuted because of a religious belief. That is an extreme case, but it shows the need to give up the absolutist view. It also shows the need to discuss particular cases to distinguish those in which dishonesty may be justified from those in which it is not.
In our discussion of each case, we first try to clarify whether tailoring the chart is a form of dishonesty and deception. If it is, we consider whether the use of deception is ethically justified. We employ a simple but useful scheme of ethical analysis. In each case, we consider the reasons in favor of deception, the reasons against the use of deception, and the possibility of truthful alternatives (1).
Mrs. P admitted herself to the hospital because she was afraid she might kill herself. She was suffering from a major depressive episode, but she improved markedly during the first weeks on Dr. A's ward. Although Dr. A believed that Mrs. P was no longer suicidal, he thought she would benefit greatly from continued hospitalization. Because he knew that Mrs. P could not afford to pay for hospitalization and that the insurance company would pay only if the patient was suicidally depressed, he decided not to document Mrs. P's improvement. He noted in the chart that the "patient continues to have a risk of suicide."
This case illustrates one of the difficulties posed by cost constraints and managed care (2,3,4). Dr. A responds to this difficulty by tailoring the chart. Does he engage in a form of deception? Yes, he intentionally misleads by what he writes and by what he omits to write in the chart. Although what he writes is true in some literal sense, his statement is misleading in the context of treatment. Mrs. P is not suicidally depressed in the way she was.
What Dr. A omits from the chart is also deceptive. Whether a particular omission is deceptive depends, in part, on the roles and expectations of the people involved. If you don't tell your colleague that you dislike his tie, that omission is not a deception. It is simply tact, unless your role or relationship involves the expectation that you offer your candid opinion. Dr. A's case is different. His professional role is to document the patient's course, and the expectation is that he will note any significant improvement. Thus his failure to accurately document Mrs. P's progress is a kind of deception.
The second and more difficult question is whether deception is justified in this case. The answer to that question depends on the reasons for the deception, the reasons against it, and the alternatives available (1,5).
The reasons for this deception are obvious. Dr. A's aim and primary obligation is to help the patient (6). He believes that Mrs. P would benefit greatly from continued hospitalization; however, she cannot afford it. He may also believe that it is unfair for the insurance company to refuse to pay for inpatient treatment of nonsuicidal depression, and that his deception rectifies that unfair practice.
There are also important reasons against this deception. The first reason concerns honesty and social trust. It is a good thing if people can rely on what others say and write. Without some degree of honesty and trust, many social exchanges and practices would be impossible (7). Deception, even for beneficent purposes, has real potential to damage social trust (1). A risk exists that deception may damage people's trust in the profession of psychiatry, and even patients' trust in their psychiatrists. Damage to trust may, in turn, compromise treatment.
The second reason concerns future medical treatment. If Mrs. P seeks medical treatment in the future, the physicians who attend to her will read the misleading notes. If they believe the notes are an accurate account of the previous treatment, they may suggest an inappropriate treatment for the present problem. Even if they have doubts about the accuracy of the notes in her chart, they are deprived of a very useful tool: an accurate history and report. In either case, the prior deception can hinder treatment.
The third reason concerns obligations and coverage policies (8,9). Dr. A seems to ignore the obligation he has to the population that is covered by the insurance policy (10). He shifts a burden onto this population by forcing the insurance company to pay for treatment it did not agree to cover. Perhaps the insurance company should pay for inpatient treatment in cases like Mrs. P's; perhaps its policies are unreasonable and unfair. But Dr. A's deception does not challenge the insurance company and pressure it to change its policy. Nor does his deception encourage patients and their families to contest the company's policies. The use of deception simply circumvents, in an ad hoc way, a policy that should be challenged and discussed.
Dr. A also seems to ignore his obligation to future patients. By introducing an inaccuracy into the chart, he compromises the value of medical records research. His deception works, in a small way, to deprive future patients of the benefit of research that relies on medical records.
Whether the deception is justified depends not only on the weight of the reasons for and against the deception but also on the alternatives that are available. One alternative is to tailor the chart. Another alternative is to describe accurately Mrs. P's response and to discharge her to outpatient care. But a third alternative exists. Dr. A can accurately document the patient's course and recommend continued hospitalization. He can petition the insurance company for coverage. If the insurance company decides not to approve further inpatient care for the patient, Dr. A can appeal that decision. This alternative is more time consuming, and there is no guarantee it will succeed, but it avoids all the problems associated with the use of deception.
Mr. Q, a patient with chronic paranoid schizophrenia, was admitted involuntarily to Dr. B's ward for treatment of acutely psychotic behavior. After three weeks, he was relatively stable and posed little risk to himself or others. Dr. B believed that Mr. Q would benefit from continued hospitalization, but she did not have grounds to keep him involuntarily. Because Mr. Q would not agree to stay voluntarily, Dr. B decided to discharge him and refer him to a community mental health center for outpatient treatment. To protect herself and the hospital against liability, she noted in the chart that the patient was completely stable and completely free of suicidal and homicidal tendencies.
Is tailoring the chart in this case a form of deception? Yes. Dr. B quite clearly intends to mislead people who might review the chart if Mr. Q harms himself or others. Her note is a form of duplicity. Although she believes that Mr. Q is relatively stable, she notes that he is completely stable. Although she believes that he poses some risk (but not enough to keep involuntarily), she notes that he is completely free of suicidal and homicidal tendencies.
Is the deception in this case justified? It is important to note that this case is quite different from a case in which a physician alters the chart to cover up negligence. Dr. B has not acted negligently. We may suppose that she has met the standards of care in evaluating, treating, and discharging Mr. Q. Why then has she resorted to deception? Physicians are sued not only when adverse events occur due to negligence. They are sometimes sued when adverse outcomes are not due to negligence, and even when there are no adverse outcomes (11). Dr. B wants to protect herself and the hospital from unfair malpractice claims.
What reasons count against the use of deception in this case? Dr. B's deception may erode trust and create confusion. Although she notes in the chart that Mr. Q is completely stable, she must voice a different message to the patient and his family. The discrepancy between these two messages may confuse anyone who receives both.
Dr. B's duplicity may also make the work of other psychiatrists more difficult. When the community psychiatrists interview and examine Mr. Q, they will probably find that he is not completely stable and completely free of suicidal and homicidal tendencies. They are then left to consider the discrepancy between their findings and the discharge summary. They may not know whether Dr. B was protecting herself, whether she misjudged Mr. Q, or whether he decompensated after discharge.
Dr. B wants to avoid being sued in cases in which no negligence exists on her part, but her practice of deception also reduces the chance that she will be sued in cases in which she is negligent. Indeed, her practice tends to pre-empt all review that depends on the chart. Perhaps malpractice litigation is a poor form of treatment review (12), but some forms of review are needed, and these forms must be based on accurate information.
What is the truthful alternative in this case? Although Mr. Q does not meet the criteria for further involuntary hospitalization, Dr. B does not need to pretend that he is completely recovered. She can accurately note his condition and note the risk that he may decompensate and pose a greater danger in the future. She can then devise and document a plan to reduce the risk of decompensation. She will need to inform Mr. Q, his family, and the community health workers of the plan, and she will need to document these conversations. If she does all this, she will have acted honestly to help the patient and to protect herself against a legal judgment. Of course, the truthful alternative does not guarantee that she will not be sued. The only way to eliminate all possibility of being sued is to stop practicing psychiatry.
Ms. R, a patient with chronic schizophrenia, became acutely psychotic after she stopped taking her medication. Her family brought her into Dr. C's office at the community mental health center. After talking to the family and interviewing the patient, Dr. C determined that Ms. R was currently unable to avoid dangers inherent in daily activities. Dr. C believed that Ms. R should be involuntarily hospitalized until her condition improved. He knew, however, that the local public hospital rarely kept patients involuntarily unless they were actively suicidal or posed an immediate and serious danger to someone else. One of Ms. R's family members had told Dr. C that Ms. R had threatened her after a long argument; however, the relative also stated that she did not feel the threat was serious. To increase the chances that Ms. R would be hospitalized, Dr. C decided to document the threat but not the extenuating circumstances.
Although what Dr. C writes in the chart is true, it is deceptive because it is intentionally misleading. To make Ms. R's threat appear more serious, he omits noting the context and the family member's reaction. He resorts to deception to benefit the patient. He believes that her impaired judgment will lead to harm, but he doubts that the hospital will hold her unless he tailors the chart.
Why does Dr. C believe the hospital psychiatrists will disagree with him? Perhaps he believes the hospital psychiatrists are not as perceptive as he is. They might focus on suicidal and homicidal tendencies and not attend to the risks due to impaired judgment. Perhaps he thinks the hospital psychiatrists overvalue liberty and undervalue welfare (13). Or perhaps he thinks they are unduly influenced by pressures at the public hospital: the limited number of beds and the workload of the staff. They may act as gatekeepers who ration psychiatric care (14). In any case, Dr. C acts to influence their judgment so they will reach a decision he believes is in the patient's best interest.
Although Dr. C acts on a desire to help the patient, there are reasons to be concerned about his deception. It may not be in Ms. R's best interest to be labeled as dangerous to others. That label has serious social consequences. And because labels tend to alter people's perceptions, Ms. R's subsequent behavior may be interpreted to fit Dr. C's deceptive note (15).
Dr. C undermines a social process that is designed to balance competing concerns and to check individual decisions. By depriving other psychiatrists of an accurate basis for evaluation, he avoids a check on his own decision. Perhaps he has overestimated the risks due to impaired judgment in this case. Perhaps he has undervalued liberty in his interpretation of the commitment statutes. Or perhaps he has ignored other concerns that should be taken into account. Maybe the hospital psychiatrists should ration care, using their limited resources for sicker and more dangerous patients. In cases of disagreement, no reason exists for society to give priority to Dr. C's judgment.
What is the truthful alternative in this case? Dr. C can write an accurate note in the chart, and then he can call the hospital psychiatrists to express his concerns. He can speak directly about the issue that may be the source of disagreement (16): the perception of risk, the balance between welfare and liberty, or the pressures to ration care. There is no guarantee that this approach will achieve the result that Dr. C deems best, but it does preserve the integrity of a process that is designed to determine whether the patient should be involuntarily hospitalized.
Mr. S, a patient with chronic paranoid schizophrenia, was hospitalized for treatment of acutely psychotic behavior. Although he had a moderate history of violence, no problems occurred during his hospitalization. After several weeks, he was stable enough to be discharged into a long-term placement. Dr. D believed it would be difficult to find a good placement for Mr. S unless she tailored his history a bit. She said that finding a placement is like selling a product: you need to present the patient in the best light. Everyone does. So Dr. D decided to write that the patient had a "minimal history of violence."
Tailoring the chart in this case is an act of deception. Dr. D intentionally alters the patient's history to achieve two goals. First, she wants to find a good placement for her patient. Second, she wants to ensure that her patient is treated fairly. If other people tailor charts and she doesn't, then her patient will be at a disadvantage.
Some serious reasons count against deception in this case. A facility might take Dr. D's description at face value and inappropriately accept Mr. S. If a facility is not prepared to deal with a patient who has a history of violence, then other patients, staff, and Mr. S himself may be at increased risk. The deception may also damage Dr. D's credibility. The staff at the facility where Mr. S is placed may learn the truth. They will then have reason to discount future recommendations from Dr. D.
The practice of tailoring the chart to place a patient is similar to the practice of inflating letters of recommendation (1). In inflated recommendations, "excellent" really means "good," and "good" means "competent." The problem with both practices is that people do not receive accurate information and do not know how to correct the information they do receive. Because everyone does not tailor charts, and because psychiatrists who tailor charts do so in different ways, scarce placements are not allocated fairly.
What is the truthful alternative in this case? Dr. D can accurately document the patient's history of violence and state the reasons why she believes he is now stable. She can follow up her written referral with a phone call to ensure that her accurate documentation is not misinterpreted, that a moderate history of violence is not taken to mean a history of frequent violence. In time, facilities may realize that Dr. D does not inflate her recommendations, that her patients are not in a worse condition than she states. The truthful alternative does not guarantee a good placement in the shortest amount of time, but it is a reasonable response to the problem of scarce placements.
We do not believe that honesty is an absolute value that always overrides all other concerns. We believe it is an important value that may, on occasion, compete with other values. Occasions may occur, even in the practice of psychiatry, when the use of deception is ethically justified. But in the four cases we considered, we do not believe that resorting to deception is justified. In these cases, the reasons that count against the deception are weighty, and the truthful alternatives are promising.
Before resorting to deception to solve a problem, it is important to reflect on reasons that may count against the use of deception. Even a seemingly simple deception may have adverse effects for the patient, other patients, the psychiatrist, other health care workers, and the profession of psychiatry. In a particular case, a deception may burden the patient with an inaccurate history or label, distort evaluation and review, damage trust in the doctor-patient relationship, affect how and when other patients are treated, undermine the credibility and character of the psychiatrist, deprive other health care workers of accurate information, damage social trust in the profession of psychiatry, or leave unchallenged serious failings in the health care system.
Before resorting to deception, it is also important to consider what the truthful alternatives are. In the four cases we considered, the truthful alternatives are relatively good. Although they do not guarantee good outcomes, they promote good care in a responsible and honest way. However, the truthful alternatives often involve more work: extra notes, letters, phone calls, and appeals. In the current health care system, there is a price to pay for honesty.
The authors thank Jean Maria Arrigo, Elena Boley, M.D., Matthew Fogg, Sylvia Hafliger, M.D., and Ping-Wen Shih, M.D.
Dr. Dwyer is clinical assistant professor of psychiatry and adjunct assistant professor of philosophy at New York University, 503 Main Building, 100 Washington Square East, New York, New York 10003 (e-mail, firstname.lastname@example.org). Dr. Shih is a resident in psychiatry at the University of California, San Francisco.