Patient privacy versus protecting the patient and the health system from harm: a case study

We report an extreme example of the relatively common problem of patients who seek care at multiple institutions or from multiple doctors. The case raises questions about risks to the patient and the staff caring for him, and specifically about patient privacy versus the potential imperative to share information with other health care providers in order to protect the patient from medical consequences of unnecessary care. A summary of some of the complex issues raised by this case is shown in Box 1.

A 71-year-old man, newly arrived in the region, presented to a rural hospital emergency department with a 1-hour history of severe, left-sided chest pain radiating to the neck and shoulder. The patient had no dyspnoea, nausea or diaphoresis. His cardiac risk factors included his sex, age, obesity (body mass index, 32 kg/m2) and hypertension (blood pressure, 148/90 mmHg). He reported a history of prior transient ischaemic attack. Cardiac enzyme levels and electrocardiograms were normal. The patient was transferred to a metropolitan centre for coronary angiography, which revealed a 30% right coronary artery stenosis with good collateral circulation. After transfer back to the regional hospital, he was prescribed cardiac medications at maximal doses and was discharged. The patient wanted to live in independent retirement housing but his family felt he was “too ill to live alone”.

Two weeks later, the man re-presented to the emergency department, again with “severe” chest pain. Diagnostic considerations included aortic dissection, recurrent pulmonary emboli, oesophageal spasm, gastro-oesophageal reflux, and coronary vasospasm. His chest pain was relieved to varying degrees with nitrates, morphine and antacids. Gastroscopy and computed tomography (CT) scans of the chest and abdomen showed only mild gastritis, and treatment with calcium channel blockers (for oesophageal spasm or vasospasm) was ineffective.

The patient had repeated episodes of pain during consultant ward rounds or investigations (for instance, he had a panic-like attack while in the CT scanner). He appeared to the medical and nursing staff to be highly distressed. In the intensive care unit (ICU), the consultant witnessed an episode of chest pain that occurred with elevated blood pressure (184/102 mmHg) and pulse (105 beats/min). Results of repeated examinations and investigations were normal. The dramatic nature of the patient’s symptoms generated attention, but his symptoms could be lessened with distraction. Panic disorder and somatisation became the leading possible diagnoses. After being reassured that he had no serious cardiac or gastric illnesses, the patient agreed that anxiety was a factor in his chest pain. Treatment with a serotonin reuptake inhibitor appeared to decrease his “attacks”, and he was discharged to live with his relatives.

The day after discharge, the patient presented with chest pain to a local general practitioner, who examined and reassured him. Three days later, his family brought him to the hospital after an unwitnessed “collapse”. The patient said that he had fallen and hit his head, stating “I’m sure I’ve had another stroke. I can’t move my left side and my right hand is numb”. On physical examination, no bruises or injuries were evident. Neurological examination revealed normal tone, no weakness, and downgoing plantar reflexes. CT and magnetic resonance imaging (MRI) brain scans were unremarkable. The next morning, after stating that he had lost vision in one eye, and with examination seemingly confirming blindness to direct confrontation, the consultant asked him to perform finger-to-nose testing with the good eye covered. He was able to do so with only inconsistent past-pointing. Sensory findings were also inconsistent: he reported loss of pain sensation on the right side, but temperature sensation on the right side was reported as “hot”. The patient’s “stroke” symptoms resolved after reassurance.

Importantly, repeated enquiries regarding his psychiatric history revealed “one episode of depression” for which he had been treated 5 years earlier at a metropolitan hospital. Follow-up of records from a former GP of the patient revealed multiple undisclosed psychiatric hospital admissions and multiple coronary care and ICU admissions across Australia, with at least 10 recent coronary angiograms (including two angiograms in different capital cities within 1 week). After initial denials regarding this history, the patient reluctantly agreed with the need for specialist psychiatric assessment. During psychiatric evaluation, the patient disclosed plans to relocate to another region rather than returning “home” to his relatives.

During this admission, the patient’s extended family formally complained to the Health Services Commissioner that the patient was being inappropriately treated and repeatedly discharged without resolution of the life-threatening cardiovascular condition they believed he had. However, privacy issues limited the treating team’s ability to speak frankly with the family and deal appropriately with their concerns. The family was informed that further discussions about the patient’s care would require his permission, which was not initially granted.

Clues to the diagnosis of this patient with multiple dramatic presentations over several weeks at one hospital for chest pain, collapse and hemiparesis included his distractibility during chest pain episodes, anatomically inconsistent neurological examination results, and his lack of injuries following a “collapse”.

Possible diagnoses included somatoform disorder, Munchausen syndrome (a severe form of factitious disorder), and malingering.1 Somatoform disorders result from the usually unconscious expression of psychological distress as physical symptoms, with the patient seeking care for multiple physical symptoms.2 In contrast, patients with factitious disorder consciously pretend to be ill or physically make themselves ill or appear ill (eg, by self-inflicting wounds or self-administering medication). Munchausen syndrome (named for Baron Münchhausen, who fabricated exaggerated stories for amusement), also sometimes called “pseudologia fantastica”, occurs in a subgroup of these patients, who feign disease, move from hospital to hospital (often travelling extensively to do so), and submit to repeated procedures for illness they have voluntarily manufactured.3,4 It is more common in women and health care workers. Malingering involves intentionally faking symptoms for secondary gain, usually in the setting of disability claims or legal battles.4-6

Our patient met criteria for Munchausen syndrome by faking symptoms of chest pain (“cardiopathia fantastica”) or stroke (“neurologica diabolica”)7 at multiple hospitals and by obtaining repeated medical procedures. Despite deliberately withholding information from the medical team and once taking money from his wife for travel to another hospital, it is arguable whether malingering was a dominant issue, as anxiety and the need for attention appeared to be more significant drivers than secondary gain. Psychiatric consultation confirmed a diagnosis of Munchausen syndrome, with elements of malingering on several occasions (eg, he “once faked chest pain while travelling” to get admitted to hospital for meals and a bed).

Somatoform disorders are common and under-recognised in primary care settings. “Non-organic” episodes may account for about half of all presentations by somatisers and patients with factitious disorder, with half related to other legitimate health issues.8,9 In a recent Australian study, 18.5% of more than 10 000 general practice patients were classified as somatisers.10 While the prevalence of factitious disorder is likely to be less than 1%, it may be higher in hospitals and specialty clinics (eg, allergy clinics).7 Concerningly, few patients with factitious disorder seek or follow through with psychiatric care.4 However, such patients can also develop “organic” illnesses, and the challenge for medical practitioners is to provide appropriate care for patients with diagnoses of somatoform or factitious disorders. Practical recommendations regarding the care of these patients are shown in Box 2.

The patient reported here in particular, and the health system in general, might theoretically benefit from an alert system that would electronically notify hospitals about his condition and the risk of performing repeated invasive investigations. However, if such a system were in place, it is possible that this patient might have organic symptoms that would be ignored or under-investigated because they are assumed to be non-organic. The status quo is that no such system is in place and the patient’s privacy is protected, at risk to himself and expense to the health system.

Although sharing of information about patients within a particular hospital or between a hospital and a treating private specialist or GP is common, the current Medical Practitioners Board of Victoria Medico-legal guidelines specify that “relevant medical staff within a hospital” may discuss a case and that “an organisation or practitioner may also use information on their files, to the extent that it is relevant to subsequent treatment”.12 In addition:

Cardiac catheterisation, the most risky of the repeated procedures in this case, carries a risk of serious complications of one in 500 to one in 1000.13 The risks include arrhythmias, cardiac tamponade, local pain, infection or haematoma at the catheter site, serious haemorrhage, hypotension, contrast dye allergy or renal injury (almost 40% in patients with underlying renal disease or diabetes13), stroke, and myocardial ischaemia or infarction. Routine risks of unnecessary hospitalisation include those due to repeated phlebotomy, repeated intravenous catheter insertion, exposure to hospital pathogens, and medication errors or side effects (eg, with glyceryl trinitrate and pain medications, including opiates).

The risk of maintaining the status quo (ie, non-notification) to “public welfare” and the health system is mainly financial, but is not insignificant. The cost to the hospital for the episodes detailed in this article alone was $32 250 during 1 month. A conservative estimate of clinical resource consumption by this patient for his repeated emergency and hospital admissions across Australia, with hundreds of ICU and medical bed-days, would include about 30 admissions with CT scans, MRI scans, and angiography at multiple centres, plus outpatient and general practice care. Thus, the cost of care for this patient’s Munchausen syndrome might easily have approached $1 million to date. Patients with somatisation, factitious disorders or malingering would consume millions of health care dollars across Australia. Inevitably, the care consumed by these patients delays care or uses resources needed for other patients with genuine medical problems.

The hospital sought a legal opinion regarding notifying other area hospitals about this patient’s potential risk of re-attendance, and was advised that the only way to provide such advice would be with the patient’s permission (eg, the treating doctors could provide a letter for the patient to take with him). However, given the patient’s lack of disclosure, it is likely that he would not give the letter to other hospitals or doctors and that he and the system remain at high risk.

Additionally, the stress placed on the treating team and staff caused by confrontations with the patient’s family over provision of “appropriate care” while attempting to protect the patient’s privacy, as well as fears relating to the potential medicolegal consequences of the family’s complaints to state officials, cannot be underestimated.14 The team is aware of at least two subsequent non-organic emergency metropolitan hospitalisations (with one cardiac catheterisation) for this patient, despite targeted GP and specialist care. However, he has also had one organic ICU admission for a myocardial infarction.

Importantly, because of concerns about “labelling” and stigmatisation of patients, privacy and legal issues, and the social and health consequences related to somatoform spectrum disorders, significant changes to the diagnostic criteria and their usage are being proposed in order to promote better medical care of such patients, specifically by reflecting the impact of external stressors on the expression of stress as physical symptoms.15

It may be helpful to consider “Project STOP”, a recently implemented program for notification of pseudoephedrine diversion.11 In this situation, the public is at serious risk from medication diversion and potential misuse or conversion of pseudoephedrine in the manufacture of methamphetamine. Under Project STOP, pharmacists in Australia may request photo identification from a client wanting to purchase a pseudoephedrine-based product; the identification details and the type and quantity of the requested product are then entered into an online database and checked against information submitted by other pharmacies. The pharmacist will be immediately notified if the client has previously purchased any pseudoephedrine-based products within the appropriate threshold period, and can make an informed decision about the sale based on the client’s therapeutic history and needs. How does this situation differ from that of patients who are “addicted to” or “misusing” medical services?

This case study illustrates a situation in which a carefully designed, confidential notification program might benefit both high-risk patients and public welfare (health system costs). More generally, assuming that confidentiality and medical care of the patient remain central to all doctor–patient relationships, the controversial issue of a national or discoverable portable medical record might increase safe, high-quality care for the patient described here, while reducing high-cost investigations for him and many others like him.

Further comments on the issues raised by this case, particularly in terms of psychiatric assessment of patients and the ethical and legal dimensions of disclosure, are provided by a psychiatrist and a bioethicist in Box 3 and Box 4, respectively.