Article Text

Download PDFPDF
Paper
To lie or not to lie: resident physician attitudes about the use of deception in clinical practice
  1. Jo P Everett1,
  2. Clifford A Walters2,
  3. Debra L Stottlemyer3,
  4. Curtis A Knight1,
  5. Andrew A Oppenberg4,
  6. Robert D Orr5
  1. 1Loma Linda University, California, Loma Linda, CA, USA
  2. 2Department of Obstetrics and Gynecology, Loma Linda University, CA, USA
  3. 3Department of Internal Medicine, Loma Linda University, CA, USA
  4. 4Risk Management and Patient Safety, Glendale Memorial Hospital and Health Center, California, CA, USA
  5. 5Fletcher Allen Health Care, University of Vermont College of Medicine, Burlington, Vermont, USA
  1. Correspondence to Jo P Everett, Chan Shun Pavilion, 11175 Campus Street,Loma Linda, CA 92350 USA; joeverett7{at}gmail.com

Abstract

Background Physicians face competing values of truth-telling and beneficence when deception may be employed in patient care. The purposes of this study were to assess resident physicians' attitudes towards lying, explore lie types and reported reasons for lying.

Method After obtaining institutional review board review (OSR# 58013) and receiving exempt status, posts written by Loma Linda University resident physicians in response to forum questions in required online courses were collected from 2002 to 2007. Responses were blinded and manually coded by two investigators using NVivo software. Qualitative and quantitative analyses of the data were performed with links to various attributes. A 95% binomial proportion CI was used to analyse the attribute data.

Results The study found that the majority of residents (90.3%) would disclose the truth about medical errors. Similarly, many residents (55.7%) would disclose the truth regarding unanticipated events, especially if the error was serious enough to result in a malpractice suit (74.7%). However, many residents (40.9%) would not reveal a near miss event because they believe it has no impact on patient health. Some residents (47.3%) would deceive the insurance company for additional patient benefits. Of those willing to lie, only a small group (4.2%) gave self-serving reasons.

Conclusions This study demonstrates that the ethical issues related to deception that trouble attending physicians also exist at the resident physician level. Residents primarily lie for altruistic reasons and rarely for egoistic or self-serving purposes that may or may not result in harm to patients, insurance companies and/or physicians themselves.

  • Truth disclosure

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Truth-telling is not a new issue in medicine. Hippocrates warned of telling the patient the nature of their illness, ‘for many patients through this cause have taken a turn for the worse.’1 However, he admonished that the truth should be told, but to a third party instead of the patient. In 1903 Cabot analysed the professional ethics of truth and falsehood, writing ‘The lies that the medical profession agree in condemning whenever the question arises are those told for personal and private gain.’2 This historic conclusion was justified by utilitarian reasoning and was based on the paternalistic assumption that the ‘doctor knows best’. This attitude in medicine led to the concept of a ‘therapeutic privilege’, which allowed the physician to do what he or she thought was best for the patient.

In recent decades, paternalism and therapeutic privilege have fallen into disfavour, while patient autonomy has come to dominate discussions in medical ethics. More emphasis is placed on principles—lying is wrong; the patient has a right to know. This change was clearly demonstrated when Novack reported that in 1961, 90% of physicians did not reveal a fatal diagnosis to patients, but in 1977, 97% did so.3 About the same time, a layman argued for truth-telling, writing, ‘The real issue is not whether the truth should be told but whether there is a way of telling it responsibly.’4 This north American change is not necessarily the standard in all cultures, however. Pellegrino addressed cross-cultural practices of truth-telling and concluded that patient autonomy was not the universal principle, but rather respect for persons.5 So if a person expected to be shielded from the truth, his or her personal belief should be honoured.

Still, physicians are frequently confronted with the ethical issue of deceiving for either the patient's benefit or their own. Physicians may lie to insurance companies, deceive about medical errors, withhold details about diagnosis at the family's request, misinform about unanticipated events, or not disclose information about near miss events.6 Truth-telling is a foundational principle of medical ethics.7 However, using a consequential method of reasoning rather than a principle-based method, professionals find situations in which telling the truth may not be in the best interest of those involved—the patient, the physician, or the insurance company.

This ethical issue also has legal implications. Lying for the benefit of the patient in order to secure additional insurance benefits—something patients may actually desire—could easily result in charges of fraud. Alexander showed that 26% of individuals would prefer their physician to deceive the insurance company after a claim has been denied.8 Previous studies reported practising physician deception of insurance companies ranged from 11% to 57.7%.9–11

However, patients do not want to be deceived about their care. Fein et al12 described the disclosure patients desire including admission of the error, the proximate effects, and any harm caused. Lying about a medical error may save the physician from a malpractice suit, but could also result in further harm to the patient. Gallagher et al13 and Kaldjian et al14 reported that attending physicians revealed medical errors to the patient between 51% and 92% of the time, whereas minor errors were only divulged to the patient between 32% and 73% of the time.

While it may not be illegal to refrain from disclosing near misses (events that almost happened, but did not) or unanticipated events (undesirable events that were not covered in the informed consent), the argument continues as to whether the patient should be informed or whether the information would only serve to increase patient anxiety. Even so, many attending physicians (73%) disclose near miss events.15

The purposes of the current study were to assess resident physicians' attitudes towards lying, compare with published data, and explore the types of lies identified by resident physicians as well as their rationalisations for lying in health care.

Methods

First and second-year resident physicians from 22 specialties at Loma Linda University (LLU) Medical Center participated in a required online programme. These courses were taught each year from 2002 to 2007 to assist in meeting the Accreditation Council for Graduate Medical Education requirement to demonstrate core competencies.16 This online series of courses entitled the Graduate Medical Education core curriculum programme, taught on a Blackboard platform, allowed residents and instructors to access the lesson material and interactively respond in an asynchronous virtual classroom. Weekly requirements included reading the lesson material, discussing the forum questions at the end of each lesson and replying substantively to at least one classmate's online message as well as all instructor-posed questions.

After obtaining LLU institutional review board review and determination of exempt status (OSR# 58013), resident responses were downloaded from three lessons that addressed deception (table 1), given a unique ID number to maintain confidentiality, and imported into NVivo, a software coding qualitative analysis program.

Table 1

Lessons related to lying

The key to these ID numbers was only available to the primary investigators to ensure unbiased coding. At least two researchers were manually involved in the coding process of resident responses. No coder developed preconceived categories, but simply allowed the posts to fall into specific branches according to content. Posts in each of the three lessons were divided into major categories based on whether residents would lie or not and then further stratified into secondary and tertiary categories based on reasons to lie.

Cases were linked to specific resident attributes including gender (56% were male and 44% were female), resident specialty and location of medical school graduation. The training programmes were divided into three groups: 56% were medicine oriented (internal medicine, family medicine, psychiatry, etc); 20% were surgery oriented (general surgery, obstetrics, urology, etc); and 24% were hospital oriented (anaesthesiology, emergency medicine, radiology, etc). The medical school locations were divided into three categories: international graduates (17%), LLU graduates (36%), and non-LLU US graduates (47%).

Coded descriptive information was listed by number and percentage. Statistical analysis was performed using the 95% binomial proportion CI to compare the difference among the resident attributes. After determining that a resident was committed to a specific response, all additional supporting responses in the same lesson were eliminated from statistical analysis in order to use the number of residents and not the number of responses as the denominator when calculating percentages.

In order to categorise the types of lies identified in coding, a scale of lies was created based in part on the work of St Augustine (box 1).17

Box 1

St Augustine scale of lies in decreasing order of severity

Lies in religious teaching (not applicable to current study)

Lies that harm others and help no one (no data fell here)

Lies that harm others and help someone

  • Lies that harm individual and directly benefit ones self (eg, doctor lies to patient to decrease malpractice risk)

  • Lies that harm and help same individual (eg, doctor gives medication to create symptoms to justify tests)

  • Lies that harm an individual and indirectly benefit ones self and do not help patient (eg, doctor withholds harmful unanticipated event from patient)

  • Lies that harm an individual and indirectly benefit ones self but still help patient (eg, doctor lies to patient about medical error)

  • Lies that harm an organisation and provide benefit to ones self (eg, doctor lies for reimbursement)

  • Lies that harm an organisation and provides benefit to others but at their own detriment (eg, doctor coaches patient regarding symptoms to justify tests)

  • Lies that harm an organisation and provide benefit to others but with poor intentions (eg, doctor maliciously deceives insurance for patient benefit)

  • Lies that harm an organisation and provide benefit to others but with mixed intentions (eg, doctor lies to colleagues for patient benefit)

  • Lies that harm an organisation and provide benefit to others indiscriminately (eg, doctor usually lies for patient benefit)

  • Lies that harm an organisation and provide benefit to others judiciously (eg, doctor lies for patient only as a last resort)

  • Lies that harm an organisation and provide benefit to others and are requested by others (eg, patient requests doctor to lie to insurance for benefits)

Lies told for the pleasure of lying (not applicable to current study)

Lies told to please others in smooth discourse (not applicable to current study)

Lies that harm no one and help someone

  • Lies initiated by doctor that have potential to harm in the future (eg, doctor withholds benign unanticipated event from patient)

  • Lies initiated by doctor that do not harm (eg, doctor withholds near miss from patient)

  • Lies initiated by others who may know the situation better (eg, family requests doctor to lie to patient)

Lies that harm no one and that save someone's life (no data fell here)

Lies that harm no one and that save someone's purity (not applicable to current study)

Bold headings are from St Augustine.

The scale was expanded to include our data followed by lie examples that were coded from resident responses. Not all of St Augustine's categories fit with the lie examples provided by residents.

Results

The categories of deception that were identified in the coding process are shown in table 2. The varying number of residents in each category correlates with the number participating in each lesson.

Table 2

Resident responses to lying scenarios

Deception of health insurance companies

Out of 636 residents who responded to the forum question related to deceiving insurance companies, approximately half (47.3%) would consider lying to the health insurance companies. The other half (51.3%) report they would tell the truth to the insurance company. Residents who graduated from international medical schools (32%, 95% CI 0.23 to 0.40) are significantly less likely to lie to insurance companies than US (49%, 95% CI 0.43 to 0.55) or LLU (51%, 95% CI 0.45 to 0.58) graduates. Residents in surgical fields (58%, 95% CI 0.49 to 0.66) are significantly more willing to lie to insurance companies compared with those in medicine-related (43%, 95% CI 0.37 to 0.48) or hospital-related (48%, 95% CI 0.40 to 0.56) specialties.

Deception about medical errors

Out of 774 residents who responded to the question relating to medical errors, only a few (2.8%) would lie or withhold information from the patient, while a majority (90.3%) reported they would tell the patient the truth. Smaller groups of residents pose questions and hesitations about telling the truth (2.5%) or would defer to an attending physician (4.5%). US graduates (93%, 95% CI 0.90 to 0.95) are significantly more willing to admit an error than international (83%, 95% CI 0.77 to 0.89) graduates, but international graduates (11%, 95% CI 0.06 to 0.16) are significantly more willing to defer to an attending physician than US (3%, 95% CI 0.01 to 0.04) or LLU (3%, 95% CI 0.01 to 0.05) graduates. LLU graduates (90%, 95% CI 0.87 to 0.94) fell inbetween US and international graduates in their willingness to admit an error.

Deception about unanticipated events

Out of 672 residents who answered the forum question relating to disclosing unanticipated events, a minority (3.3%) stated they would not disclose unanticipated events to patients. A substantial group (41.1%) would not disclose the information unless specific circumstances overrode their reticence, such as importance to patient health or being directly asked by the patient. Over half (55.7%) reported they would always disclose unanticipated events to patients.

US (51%, 98% CI 0.44 to 0.57) and LLU (46%, 98% CI 0.39 to 0.54) graduates are significantly less likely than international graduates (71%, 98% CI 0.62 to 0.80) to disclose an unanticipated event. Residents in hospital-based specialties (42%, 98% CI 0.33 to 0.50) report unanticipated events significantly less than those in medicine-related (57%, 98% CI 0.51 to 0.63) specialties. Surgical residents (53%, 95% CI 0.43 to 0.62) tend to disclose more than those in hospital-based specialties (42%, 95% CI 0.33 to 0.50).

Disclosure about unanticipated events is the only category of deception in which gender is significant. Male residents (46%, 95% CI 0.42 to 0.51) report unanticipated events significantly less than female residents (60%, 95% CI 0.55 to 0.66). However, male residents (43%, 95% CI 0.39 to 0.48) choose to disclose under certain circumstances (eg, if the patient asks) significantly more than female residents (32%, 95% CI 0.27 to 0.38).

Deception about near miss events

Out of 492 residents who answered the question regarding disclosing near miss events, over one-third (40.9%) reported they would not reveal the near miss. A smaller group (33.3%) indicated they would always disclose the near miss. Smaller still was the number of residents (25.8%) who would only disclose under certain circumstances (importance to patient health). None of the attributes examined were significant.

Types of lies

Figure 1 depicts the characteristics of lies described by resident physicians that depart from the truth in four directions: altruistic lies, egoistic lies, and lies either resulting in harm or no harm. Whereas altruistic lies are those that provide benefit to the patient, egoistic lies are those that provide benefit to the physician. Lies that cause harm refer to any harm done to patients, insurance companies, or even the physician. The table depicts possible combinations of lie characteristics. For instance, a physician exaggerates symptoms (harming the insurance company) in order to secure benefits for the patient (altruistic towards patient). The placement of these points on the graph was based on St Augustine's (box 1) continuum of lies that identified degrees of harm as well as who benefits from the lie.17 Following St Augustine we placed the most serious lies (those that harm others and benefit oneself) and the least serious lies (those that harm no one and provide benefit to another) at opposite corners of the graph. The remainder of the lie categories was placed on a continuum within the four quadrants based on the degree of lie seriousness as assessed by the authors. The size of each point graphically represents the percentage of residents willing to deceive as found in our study.

Reasons to lie

Table 3 outlines the reasons that residents choose to lie in each of the scenarios.

Table 3

Resident reported reasons to lie

When lying to the insurance company the most common reason is obtaining treatment for the patient, followed closely by wanting to ‘get back’ at the insurance companies. In near miss situations, residents stated the reason not to disclose was that the event did not take place and was most likely unimportant to the patient's health. Patient health is also the foremost reason for non-disclosure in an unanticipated event. Finally, those few residents who chose to deceive about a medical error most commonly do so because they prefer not to admit that they made a mistake. Of the responses indicating willingness to lie, only a small group of residents' lies (4.2%) were self-serving.

Conclusions

Many healthcare professionals have their own definitions of types of lies or are familiar with those described by Bok18 including white lies, excuses, lies in crisis, lies to protect others, and lies for the public good. Bok further adds that ‘some consider all well-intentioned lies, however momentous, to be white’. Bok describes her personal definition of the white lie as ‘a falsehood not meant to injure anyone, and of little moral import’.18 In our study, residents often justify lying as long as the patient is benefitted as pointed out by this resident, ‘I think the wording of LIE involves that we are out for our own interest and not the patients'. When in reality we are all about patient care.’

The residents themselves said that they are attempting to behave altruistically, or as Darwin19 put it, ‘incurring a personal cost that in turn benefits others’. Darwin's favourite example of altruism was honey bees. He found that in a hive there are sterile bees that do not reproduce but rather supply resources to those bees that do reproduce. These same bees will also defend the hive with their lives. While our residents are not giving their lives to gain additional insurance benefits for their patients, they are behaving altruistically.

When lying for the patient, the insurance company loses money, the patient may receive tests of questionable necessity with potential complications, and the physician is sacrificing moral integrity. However, deception for the benefit of the patient may be considered a form of benevolent deception, a concept described by Jonsen,7 but it is still a lie. In contrast, the insurance industry has a different view of lying that resides with a legal definition. The False Claims Act states any person who presents a false claim or makes a false record is liable pay a fine of at least US$5000 in addition to three times the amount of damages sustained.20

Occasionally, patients may actually approve of physician deception. A 2003 study by Alexander et al8 of 700 individuals showed that 50% stated it is acceptable to misrepresent facts and 41% stated it is acceptable to lie for the benefit of the patient. Moreover, 19% indicated that a physician had used deception on their behalf in the past. Our study showed that patients requested their doctors to lie to the insurance companies on their behalf, representing almost 6% of all types of lies reported.

Certain practices of insurance companies may be cause for some physicians to lie in retaliation. The California Medical Association reported three large suits against Health Net, Blue Cross and Wellpoint/Anthem within the past 4 years.21–23 In each of these suits, the companies were charged with a variety of actions including promising medical coverage, then dropping patients if they needed expensive treatment, engaging in practices that led to denials of claims, increasing premium charges to customers without advance notification, refusing to pay for patient care after pre-authorising the care, reducing reimbursement levels without appropriate notice and revoking patient health plans retroactively. The study in 2002 by Werner et al9 demonstrated that attending physicians are more likely to use deception with insurance companies if the denied claim appeals process involved more time or was associated with less success. Residents' animosity towards the insurance companies is also demonstrated in this sample post, ‘Insurances are here in business. They are not here for your patients’ well-being. If I think a test is necessary for my patient, then I will get it. I wouldn't accept a medically illiterate person to tell me what to do.' Our study also demonstrated some resident attitudes about the inequality of the lying game when physicians are expected to behave ethically with regard to insurance companies, while the insurance companies act in their own interests even to the perceived detriment of patient care.

Resident physicians are thus faced with balancing competing values—to maintain personal and professional integrity by doing what is ethically right, to do what they believe to be in the best interests of the patient, to do what is required by law, and to avoid harm. Some residents link the lying issue to legal requirements as this resident responds, ‘It's not really an argument about ethics. It's an argument about legality, a totally separate issue. If we are legally required to disclose this, we do it. The law trumps all.’ The American Medical Association also addresses the issue of legality in its Code of Ethics: ‘Ethical values and legal principles are usually closely related, but ethical obligations typically exceed legal duties. In some cases, the law mandates unethical conduct. In general, when physicians believe a law is unjust, they should work to change the law. In exceptional circumstances of unjust laws, ethical responsibilities should supersede legal obligations.’24

When we asked residents about disclosing unanticipated events, we did not specify the event, but allowed them to come to their own conclusions. Over half (56%) chose to disclose the event. When we later asked the same residents about disclosing an unanticipated event if they knew a lawsuit would ensue, 75% chose to disclose the event. This demonstrates that the residents generally are more concerned with their patients than with their own wellbeing.

Residents' decisions are affected by many factors. International residents are less willing to lie to insurance companies about patient symptoms and to patients about unanticipated events than those who trained in the USA. However, they are less likely to report medical errors than those who trained in the USA. These are similar to results found in the study in 2004 by Lee et al,25 which showed that international graduates are less likely to change the patient's official diagnosis than US graduates. Does this mean that US medical schools are not addressing the ethics of lying or alternatively that international medical schools place less emphasis on doing what is best for the patient, whatever the cost?

A study by Garbutt et al15 in 2008 showed that surgeons are less likely to report medical errors. We found no significant difference between departments with regard to disclosing medical errors; however, significant differences do exist in lying to insurance companies and disclosing unanticipated events. Surgical residents are more willing to lie to insurance companies, and hospital-based residents are more willing to deceive about unanticipated events. Perhaps the surgeon is more focused on the mission to be accomplished and less concerned with the means used, whereas the non-surgical physician may be more contemplative regarding the consequences of how one arrives at the altruistic value. Is it possible that hospital-based residents have more experience with procedures and therefore encounter unanticipated events more often than those who are medically based?

A 1991 study by Roter et al26 showed that female physicians took more time with their patients, encouraged more patient–physician partnering, and gave more information during clinic visits than male physicians. Our finding that female physicians were more willing than male physicians to disclose unanticipated events may involve this willingness to spend more time to give patients information while encouraging them to be more proactive in their own health care.

A limitation of this study, as with all qualitative research, is that the coding process remains subjective even when two coders are involved. In addition, what residents report they would do while writing in an online classroom may not directly correlate with what they would do in reality. Having a multicentre survey might strengthen the breadth of the deception practices in the general population of resident physicians.

In practice no physician contemplates engaging in any dishonest behaviour. Such actions clearly go against the Hippocratic oath, but potentially engaging in deception confronts physicians in not uncommon situations, as graphically shown in the types of lies diagram (see figure 1). Doctors communicate in different ways that hopefully help patients understand their clinical challenges. A doctor may not tell all that is known because the patient cannot handle or understand the information. Doctors are careful not to take away hope when patients ask about impending death. Sometimes the question is not answered, but the patient is told that ‘we are going to take good care of you’. Becoming dishonest that involves legal issues such as insurance deception for the benefit of the patient or the doctor, or withholding or lying about information to protect a physician from a lawsuit clearly are more egregious forms of deception, some of which border on fraud and should clearly be avoided.

Our study demonstrates that the ethical issues related to deception that attending physicians face also exist at the resident physician level. Residents are trying to learn how to be professionals, but are also faced with concerns about deception. While the majority will disclose a medical error to the patient, only half will be truthful with the insurance company, and only one-third will consistently disclose near misses to their patients. Over half will disclose an unanticipated event to the patient, but in making a distinction between benefitting the patient and protecting themselves, three-quarters of the residents will disclose, even if such disclosure results in a lawsuit. Whereas our formal curriculum challenged the resident to consider the issues related to deception, it is clear that residents are still contemplating the matter and struggle with balancing altruism, egoism and beneficence. However, the overwhelming majority embraced an altruistic rationale regarding deception. Finally, residents may benefit from a formal educational curriculum to achieve a confident understanding regarding the ethical virtue of truth-telling in the current complex world of health care.

References

Footnotes

  • Funding Funding was internal and provided by the GME Core Curriculum Program.

  • Competing interests None.

  • Ethics approval This study was conducted with the approval of the institutional review board, OSR# 58013.

  • Provenance and peer review Not commissioned; externally peer reviewed.