Practical tips for helping patients and simultaneously avoiding legal battles.
SPECIAL REPORT: FORENSIC PSYCHIATRY
Many psychiatrists are sued for malpractice every year. Some deserve it; some do not. We offer some practical tips for helping patients (most important), decreasing the chance of being sued, and increasing the odds of winning if you are. These are not legal tricks, but practice errors that plaintiffs’ lawyers look for, and ways to correct them.
To be successful in a malpractice action, the plaintiff must prove 4 things (Table). If any of those 4 elements is not proved, then no malpractice has occurred. The level of proof required is fairly low, compared with criminal matters: preponderance of the evidence (merely more likely than not).
The authors have been involved in the clinical, legal, and administrative aspects of psychiatric malpractice cases, particularly those related to suicide, for more than 3 decades and in more than 30 states. We see the importance, to patients and courts alike, of practicing well. Neither medicine nor the law asks for perfection, but both demand competence and mindfulness of the concepts we will illustrate. Our comments are not meant to preach about clinical practice—many readers should already know most or all of what we describe—but to alert clinicians to the things that influence lawyers and juries as they consider alleged malpractice.
Most lists of the reasons people sue for psychiatric malpractice are misleading. The great majority of such lawsuits are related to one thing: suicide. Many apparently non-suicide topics on those lists—such as inadequate assessment, treatment or follow-up—refer to cases in which suicide occurred. Others, such as violating confidentiality, lack of informed consent, failure to monitor lab results, adverse medication reactions, and danger to others are far less common than readers might think. Another, sex with patients, is almost always adjudicated as a nonmalpractice damage issue or a criminal and/or licensure offense, rarely as “malpractice” (and rarely covered by malpractice insurance).
Finally, some of the examples below allude to a combination of clinician and hospital responsibility. It is important to note that the standard of care (SOC) in malpractice cases requires physicians to recognize, within reason, unsafe treatment environments and procedures to which their patients are exposed, to protect patients from them, to bring them to others’ attention, and to advocate for their elimination.1
Treat patients as you want your family members to be treated, and give time. Be sure to give the time needed in assessments, personal contact, family feedback, and follow-up. Take time explaining things, and time documenting in sentences rather than acronyms or quick words that do not reveal your thinking process and judgment. (More about medical record notes later.)
Understand the assessment, recognition, and management of suicide risk. Competently assess patients who are potentially suicidal, do so completely and regularly, not just on admission and discharge.
The point is risk assessment, not prediction. Suicide is unpredictable. So what? Don’t get hung up on predictability. Your job is to deal with risk. In malpractice cases, the legal test is foreseeability, defined as anticipation of risk, not predictability.
Be the doctor. In most of the case examples, both a psychiatrist and a hospital were sued. Make no mistake, psychiatrists are important to facility policies and procedures, and to the individual actions of staff with whom they work. You are a care leader. You should know whether or not your hospital or clinic and its staff are treating patients properly. Even when hospital staff or procedures are faulty, juries often view the doctor as someone who failed to protect the patient from inadequate care (and they may be right). You have an ethical and legal duty to do or order what your patient needs in order to be safe during, and because of, treatment, regardless of schedule crunches, staffing limitations, or administrative pressures.
Document your judgment. Write more, not less. Short notes are very often inadequate, and abbreviated terms such as “No SI/HI” say little to malpractice juries (and clinical readers). How did you assess suicide risk and arrive at your decisions? What was your thinking?
Malpractice liability rarely comes from honest errors in judgment; it comes from lack of adequate examination and failure to use intelligent reasoning (eg, Tkacheff v Roberts2). If there isn’t clear evidence of reasonable professional judgment in the medical record, lawyers and jurors often assume there wasn’t any. Efforts to convince them by testifying months or years later sound hollow. If there is reasonable evidence supporting your actions in the record, lawyers usually do not file suit.
So-called “positive” or “protective” risk factors are worthless for assessing individual patient risk. Loving families, good jobs, status, certain age groups, and other conditions may be statistically associated with lower group risk, but patients are individuals. Suicide occurs in all social strata, ethnic groups, religions, and age groups; in close families and broken ones; and in people with jobs and without them.
“Dr Martin,” a successful scientist at a large government laboratory, was being treated for depression and severe anxiety by his family doctor. He declined referral to a psychiatrist, fearing his security clearance and position would be affected. One evening after dinner with his wife and children, he went into his study and shot himself in the head. The suit filed against the family doctor alleged that she had not adequately assessed the patient and inappropriately allowed his status to interfere with her pursuing necessary psychiatric referral.
“Dr Boris,” a prominent physician, was hospitalized after a large overdose that he said was “accidental.” The psychiatrist who assessed him strongly recommended inpatient psychiatric evaluation. The patient refused, saying he had “learned his lesson.” His wife was adamantly against inpatient treatment as well, referring to their close family and promising “We’ll take good care of him.”
The psychiatrist asked for a second opinion. The second psychiatrist concurred with the first. An involuntary hold was accomplished and the patient was allowed to go to a distant facility for further care.
At the end of the initial protective detention (5 days), the patient convinced his inpatient psychiatrist, who was aware of the overdose but never contacted the earlier consultants, that he no longer needed hospitalization. He was discharged and died by suicide within 24 hours. Malpractice suits against the distant facility and psychiatrist were settled for an undisclosed amount.
Get consultation. Getting competent consultation before denying hospital admission, decreasing level of observation, or ordering discharge when a patient is at risk of suicide not only decreases the odds of a bad decision, it also reduces the chances of being sued if a tragedy occurs.
Don’t rely solely on the patient for information about suicidal thoughts and behaviors. If there is no other information source, be very cautious. Many patients lie or hide the truth. Even those who answer truthfully often do not understand their own symptoms, and cannot predict their future behavior.
“Mr Young” was brought to a general hospital emergency department by police officers, who had found him intoxicated and sitting precariously on an overpass railing above a busy highway. He said he was planning to jump in front of “the next 18-wheeler that comes along down there.” He was admitted for intoxication with suicide risk. Mr Young’s wife arrived as he was being transported to the psychiatric unit, but she was not allowed to accompany him. Staff added her contact information to the chart.
An inpatient psychiatrist saw him the next morning and reviewed the admission. Mr Young was anxious to go home, assuring the doctor that he had been drunk and had “acted stupid” the night before. “I was clearing my head, thinking about getting a divorce, not about jumping off the wall.” When the doctor asked about the police report, he said “I don’t remember that…they were trying to help, but I’m sober now and I don’t belong here.”
The psychiatrist went through the same suicide risk checklist that the emergency department social worker had used, relying on the patient’s own responses. Mr Young denied prior psychiatric history, suicidal thoughts, and past attempts; the checklist score was well below “high risk.” The psychiatrist ordered discharge for that morning with a note that said, “Pt. wants disch. No SI/HI apparent. No intent to die. Mar. probs./ intox… Own transp(ortation). Call PRN.”
Mr Young left the hospital via a hired car service and went to a bar. Several hours later he was struck and killed by a truck on the same highway where he had been found the prior night.
His widow sued the psychiatrist and the hospital. Past records and her deposition testimony indicated recent treatment for both substance abuse and depression, and a serious suicide attempt while intoxicated a few weeks before this incident. His wife testified that she had tried to talk with the admitting physician but was told to contact the psychiatric unit in the morning. She “thought he was safe in the hospital…[and] that they’d help him and call me and I’d tell them what he’d been doing…but they didn’t call, and when I finally called them, they’d already discharged him and nobody knew where he was.”
The attending psychiatrist settled out of court. The hospital, which employed the emergency department physician, lost at trial.
Never rely on “contracts for safety” (CFS, no harm contracts). Many hospitals and some psychiatrists still use these forms, despite the many studies that show their uselessness (and sometimes the damage they can do) for patients who are suicidal. Never rely on these forms when making decisions about admission, monitoring, or discharge.
“Mr Downton,” a widower and retired elementary school principal, shot himself after being released on pretrial bond for allegedly exposing himself to children. He was admitted; the wound was found to be non–life-threatening, and he was transferred to the psychiatry department on continuous observation.
Mr Downton slept poorly, did not participate in unit activities, declined contact with his only son, and was often seen staring tearfully out a window. He complained bitterly that being continuously observed made him feel worse, and said he had trouble using the restroom with an attendant talking to him through the door.
The nursing staff asked if his monitoring could be changed to checks every 15 minutes (Q15), citing a staffing shortage and 3 days of “No SI/HI” in the nursing notes. His psychiatrist talked briefly with Mr Downton, who denied suicidal thoughts, seemed optimistic as he spoke, and promised to tell staff if “those feelings” returned. Nursing policy required a CFS before the close monitoring could be stopped. He readily agreed and one-to-one monitoring was discontinued.
That evening, on the second Q15 round of the night shift, Mr Downton was found hanging by a sheet from his bathroom door.
The psychiatrist and the hospital were both sued. The plaintiff’s case hinged largely on expert testimony that the psychiatrist should have recognized substantial risk in Mr Downton’s history and symptoms; the suicide risk assessments were inadequate; Q15 is inadequate protection and below the standard for such patients; and it is below the SOC to rely on a CFS. The defendant psychiatrist testified that the CFS was a significant factor in eliminating continuous observation, and one that he and the staff often relied upon to decrease monitoring and “use nursing staff more wisely.” The plaintiff prevailed in both cases.
This vignette illustrates several critical errors. The patient had showed no objective signs of decreased suicide risk before 1:1 was discontinued; he had merely spent a few days without talking about suicide. Neither the psychiatrist nor the nursing staff had documented evidence of adequate suicide risk assessments after his admission. The record contained little description of the physician’s judgment in allowing release from 1:1 and deciding that Q15 was adequate for a greatly shamed man who had tried to shoot himself just a few days before. There were physical inadequacies with the patient’s room and bedding that made it easy for him to hang himself. In addition, posttrial juror questioning revealed that several faulted the physician for “going along with” a hospital CFS policy he should have known was wrong.
When considering discharge or decrease in monitoring level, be able to document what has changed for the patient. Make sure your documentation is more than a description of the time spent on the unit without an attempt. If there are no substantial, reliable changes, be very careful.
Beware of 2 substantial and common risk factors that are not on most checklists: instability and unpredictability. Because suicide is unpredictable, it is important to be cautious. Patients who are unstable, whether from psychosis, severe depression, bipolar disorder, severe anxiety, substance abuse, borderline syndromes, recent loss, or something else, are inherently at higher risk.
“Ms Reynolds,” a young woman with a history of intensive but mercurial relationships, volatile moods, and episodes of overdose and cutting her wrists, stopped her medications and was hospitalized after stabbing herself during an argument with her boyfriend. She had accused him of causing her to miscarry and then cut herself across the abdomen.
Ms Reynolds seemed to respond to medication and support over 4 days of close observation. She denied suicidal thoughts and participated in groups. She talked with her boyfriend by phone, sometimes calmly but sometimes angry that he had not visited her. When he finally came to visit, she was inappropriately affectionate.
She requested discharge on her fifth hospital day. Nursing notes said her insight, judgment, and impulse control were “normal,” based primarily on her agreeing to take medication and denying suicidal thoughts. Her psychiatrist, relying on the nursing notes and brief daily interviews, discharged her within hours of removing her from continuous observation, and without a systematic risk assessment She was given a supply of medication and a clinic appointment in 3 weeks.
Ms Reynolds was found dead from an intentional overdose a week before her outpatient appointment. In the malpractice action, the defense argued that “You can’t protect patients from themselves; you have to discharge them sooner or later,” and said her prior overdoses and cutting were not “real” suicide attempts. The case was settled before trial.
This vignette illustrates the need to keep patients safe until treatment response is well established, then carefully reassess their risk. In addition, transition to aftercare is important, easing the change from 24-hour observation and support to virtually none. Discharge planning should include contact with the outpatient caregiver before discharge and a first outpatient appointment within days, not weeks.
The vignette also illustrates the folly of calling self-destructive behavior “minor” or “gestures.” There is no such thing as a suicide “gesture.”
Fifteen-minute checks (Q15) are below the SOC for patients with even moderate suicide risk. It is easy to kill oneself in 15 minutes, or 5 minutes for that matter, even if the checks are carried out as ordered (sometimes they are not). Jayaram et al.3, describing very successful Johns Hopkins risk management procedures, are among the many who have established continuous observation as the SOC. Chassin and Loeb4 agree, citing Joint Commission on the Accreditation of Healthcare Organizations experience and a large professional literature.
“Mr Kane” was placed on Q15 for suicide risk. He was found dead during shift change early the next morning. His time of death was determined to be around 3:00 AM, hours before he was found, but the Q15 record sheet indicated that an aide had observed him sleeping every 15 minutes through the night shift. At deposition, the aide admitted falsifying the Q15 shift record, with the excuse that “We do that all the time. I check the patient every 15 minutes, then fill everything in at the end.” He couldn’t explain why he had not noticed the patient hanging in the room for hours.
In the subsequent lawsuit, lack of continuous observation was found to be the main cause of Mr Kane’s death. The psychiatrist was found liable for not ordering continuous observation. The hospital settled out of court.
Falsified Q15 records are not rare. Nevertheless, the primary lesson is that if Mr Kane had been on continuous observation, his death would almost certainly have been prevented.
Communicate with corroborating sources, including family members. We have long taught 1 way of gathering of important clinical and safety information does not require a release of information (so long as the informant knows that the patient is in your care). Here is how it works: The doctor tells the relative or other person, “I cannot share what (the patient) has told me, but I would very much like for you to answer a few questions and tell me what you can.” Relatives and friends have no duty of confidentiality to the patient.
Do not let false confidentiality concerns threaten a patient’s life or limb when they are at risk for suicide or other substantial harm.
Communicate with co-treaters. Regular contact with co-treaters is important to meeting the SOC. Unless you have terminated care entirely, you are still the patient’s doctor. Be sure that counselors and programs give you regular follow-up, document it, and do the same for them. There is no issue of clinical confidentiality between co-treaters.
Know your referral resources and co-treaters. You are responsible for knowing the general qualifications of clinicians and facilities to whom you refer. If you reasonably should have known that they were incompetent or otherwise inappropriate for the referral, you may be partially responsible for a negative outcome.
Never assume that a patient’s family can or will monitor the patient as well as hospital staff. Asking family to carry out hospital responsibilities is below the SOC. (Like several of the previous examples, this caveat applies to both hospital discharge and failure to admit an outpatient who is at substantial risk.)
After several days of psychiatric hospitalization following a hanging attempt, “Mr Frank” demanded release. His psychiatrist and the hospital discharge planner spoke with Mr Frank’s family, who promised to watch him “24/7.” Mr Frank was found hanging in their garage early on the second day after discharge to their care.
In the subsequent lawsuit, the family testified that they indeed had said they would monitor Mr Frank, and part of the psychiatrist’s defense was that he had essentially discharged Mr Frank to another “care environment”—his family home—as agreed. A jury found, however, that the psychiatrist was not entitled to rely on the family. They found that the family had no duty and no reasonable expectation of an ability to treat the patient as he would have been treated in a hospital, and releasing Mr Frank to the family was below the SOC and caused Mr Frank’s death.
Understand risk. Practice well. Document scrupulously. Be the doctor.
Dr Reid is a clinical and forensic psychiatrist who has worked with attorneys in scores of malpractice actions. He is professor of psychiatry at the University of Texas Dell Medical School and clinical professor at Texas Tech University Health Sciences Center. Mr Simpson is a malpractice attorney for psychiatric and mental health organization matters, with cases throughout the United States. He is on the Board of Directors of the American Association of Suicidology.
1. Joint Commission on the Accreditation of Healthcare Organizations (JCAHO). Suicide Reduction Measures. In: JCAHO. The Physician’s Promise: Protecting Patients from Harm. 2nd Edition. Joint Commission Resources (JCR); 2006.
2. Tkacheff v Roberts, 47 NYS 3d 782, 2017 NY Slip Op 01429, 1, 2017 WL 702943 (NY 2017).
3. Jayaram G, Sporney H, Perticone P. The utility and effectiveness of 15-minute checks in inpatient settings. Psychiatry (Edgmont). 2010;7(8):46-49.
4. Chassin MR, Loeb JM. High-reliability health care: getting there from here. Milbank Q. 2013;91(3):459-490.