February 02, 2012

Dear Dr. Jeff:
Our nursing home was devastated when a new resident shot himself on Christmas Eve. He had been admitted 2 days previously for rehabilitation after suffering a small stroke. He persuaded a friend to bring in his revolver — which he usually kept next to his bed at home — so he would feel “safe” surrounded by strangers. After the friend left, he shot himself in the head. He was an 80-year-old man who lived alone. There was no known history of depression. We checked carefully with the hospital and his primary care physician, but he had never threatened suicide or taken antidepressants. One staff member stated that she had heard him say, “I wish this was all over,” but did not consider this any kind of threat. The police and the state health department both reviewed the case and found no issues with our care. But I wonder if we could have done more.

Dr. Jeff responds:
Holidays can be sad times to be alone, and those at year’s end can be particularly difficult because we associate them with the ideal of an intact family celebrating together. Decorating trees, singing traditional songs, opening presents, and eating traditional foods are all intended to be enjoyed with those who are “near and dear to us.”

Many cultures share joyous holidays around the winter solstice, when the “rebirth of the sun,” an event that the Romans celebrated on Dec. 25, promises longer days. New Year’s is a traditional time for reflection and hopes for a better year to come.

But hope does not spring eternal. These same dates remind many seniors of loved ones who have died, of children they never had or from whom they are permanently estranged, of connections lost and opportunities wasted. We should not forget that the most popular Christmas movie, “It’s A Wonderful Life,” is about a man who plans to kill himself on Christmas Eve. Clearly, your resident lacked an angel second class hoping to get his wings. Some studies suggest that Christmas is the second most common date for suicide (Mother’s Day being first).

Although death in the nursing home is common, suicide is not. Any unanticipated death should lead the care team to examine its work, whether unofficially or through a formal morbidity-and-mortality session, as part of overall quality improvement. Often, the interdisciplinary team finds that it simply failed to recognize how sick the resident really was. This can apply to mental illnesses as well as physical diseases. Multiple studies have documented the failure of clinicians to recognize that their patients are terminally ill, even when they can recognize end-of-life status in a similar patient not under their care. Depression can be equally difficult to recognize when we are close to it. Certainly, the death you describe was unexpected and merits some review.

Suicide, however, is more than simply an unexpected event. It is always perceived, at least on some level, as a condemnation of those left behind. The survivors inevitably feel guilty. Indeed, for some people committing suicide, that is a primary goal. Many of them wish to punish their family or friends.

There is always a shock to our own sense of the world to discover that another person whom we care for, even if only professionally, does not consider life worth living. Even though your care team had a fairly brief time with this resident and was almost certainly not the target of this act, it is important to recognize these guilty feelings and provide some opportunity for your staff to discuss them. Ideally, a bereavement counselor or other mental health professional could work with the staff.

The Risk Factors

This death was not, however, a completely random or totally inexplicable event. For several reasons, this patient may have fallen into an extremely high- risk group. First, he was male. Although suicide attempts and gestures are more common among females, completed suicides are more common among men. This may be because men are much more likely to use firearms, which have a high fatality rate, while women are more likely to use pills whose effects can be reversed (narcotics), which rarely cause death (benzodiazepines), or are slowly absorbed in the stomach and so can be blocked by activated charcoal and stomach flushing.

So suicide attempts with pills are, fortunately, rarely successful. (Of course, we don’t really know whether some elderly patients whose deaths are attributed to one of their multiple medical conditions may have deliberately hastened the process by taking too many or too few of the multiple medications that have been legitimately prescribed for them.)

Second, he was old. Although the tragedy of teenagers taking their own lives receives extensive press coverage, the prevalence of suicide actually increases significantly with old age. Except for the occasional elderly couple with a suicide pact, these deaths among the old and frail rarely attract media attention. But clinicians should be aware of the true demographics.

And this resident had an additional risk factor: his recent stroke. Studies suggest that as many as 90% of stroke victims experience significant depression after the episode. Whether one wishes to adopt a neurobiologic explanation that connects the death of brain tissue to a loss of neurotransmitters or a psychosocial one that ties functional loss and illness to emotional loss and a blow to the ego, strokes appear to be strong predictors of emotional distress in elderly people, among whom depression is already common. Conversely, recovery from depression is a powerful predictor of functional recovery.

For these reasons, many stroke- rehabilitation centers have adopted routine psychological screening of all patients at admission. As the locus of poststroke recovery has shifted to skilled nursing facilities, we need to consider incorporating some of these best practices in our routine policies.

Our admission process also should include an interview focusing on vegetative signs of depression and the patient’s premorbid psychiatric status. Since a per- son’s risk is strongly associated with a family history of suicides, this area needs to be explicitly reviewed. Past suicide attempts, alcoholism, and borderline personality disorder are also major risk factors for suicide attempts and completed suicide.

Clinicians should be aware that the Public Health Questionnaire‐9 (PHQ‐9), a validated depression screen, is a required portion of the MDS 3.0 that is administered after admission and periodically throughout a resident’s stay. Because the PHQ‐9 usually is completed around day 7 after an admission, it may not be sufficient for high-risk patients. But it is certainly a long step in the right direction, particularly if mechanisms are in place to ensure that positive findings are rapidly transmitted to the attending physician and care team.

Means to the End

Another risk factor is the ready availability of means for a patient who might impulsively take his or her own life. Firearms are the most common such tools, and their availability has been documented as a suicide risk factor. Polypharmacy has encouraged the ready access to a wide variety of potentially toxic substances for many seniors. In urban areas, jumping out of a high window is possible and not uncommon.

Nursing homes have been comparatively safe because of federal and state design requirements, locked medication carts, and medications administered in unit doses under supervision. The unfortunate lack of privacy in most institutional settings carries the unanticipated benefit of closer patient supervision, if only by roommates. No one in a nursing home is truly alone.

But many seniors have spent their lives around firearms. Veterans, of course, were taught to shoot in basic training. Hunting is part of many rural cultures. When I was a child growing up in Alaska, the University of Alaska women’s dormitory had a rule that guns had to be left in the gun racks in the lobby (apparently, 1950s sexism allowed males to keep their rifles in their rooms). The prevalence of handguns is a relatively modern phenomenon.

There is an ongoing political debate regarding concealed weapons, including in states that have specifically approved their possession in churches and colleges. Those of us who practice in locales with more restrictive legislation may need to adopt policies and procedures excluding firearms from our buildings. This may be particularly important in those states that are contemplating the transfer of prison populations into long-term care facilities (see my December 2011 column, “Offenders in Long-Term Care Facilities”).

Some extreme advocates of “patient choice” might applaud your facility’s failure to subvert this resident’s choice to end his life. Although Dr. Jack Kevorkian died last year and the Hemlock Society watered down its name to Compassion & Choices, many Americans still support “the right to die.”

In three states, your resident might even have asked his physicians to assist him with a prescription, saving the cost of a bullet. However, for this man who was neither terminally ill nor suffering in any obvious way, it is difficult to see this act as anything other than an expression of mental illness running its course in unfortunate circumstances.

We cannot protect our residents from the reality that aging may be associated with declining function, the loss of independence, and deaths among friends and family. But we can treat depression, and we can try to create a community that offers stricken people a future including comfort and joy.

DR. NICHOLS is the vice president for medical services of the Cabrini Eldercare Consortium in New York City, which includes two skilled nursing facilities, three home care agencies, two adult day care programs, and a senior housing complex. He invites your questions for possible discussion in this column. Please submit them by e-mail to caring@elsevier.com.

<- Back to: Dear Dr. Jeff