Predicting Disaster

This post would have been much more topical in January of this year, but it still illustrates the worst that can happen. The lead up to the potential tragedy is something I regularly encounter, as the psychiatry resident on call.

In January of 2017, a man with PTSD went to the ER in Antigonish. It is unclear to me what happened next, but he was sent home. There was allegedly some past history of domestic abuse; how much I don’t know.

He killed his family and himself within twenty-four hours.

As far as I know these are the facts of the case. My knowledge comes from the same news stories and reports as anyone else. One CBC news article alleges that there were no beds available in the hospital. Another cautions that PTSD is likely not the sole or main factor involved. There is (or was) a review underway to determine what help was sought and what plans were made.

If he was seen by physicians that night, I am moved to pathos for them. This outcome is the worst that any Doctor would fear for his or her patient.

However. It’s extremely difficult to predict the future. This is true for every prognosis, analysis, battle plan. In psychiatry, I have heard it said informally that ‘it is always the ones you don’t expect that kill themselves.’

I’ve been the one making that call. I’ve detained people under the mental health act who want to leave, and some who don’t but who are too sick to have that option left open.  I’ve discharged people who want to go home, and some who want to stay. I’ve been yelled at, threatened with lawsuits, cursed with “if (s)he commits suicide, it’s your fault.” Even so, the plan did not change.

It’s a difficult situation. I hope it’s obvious that I want to take care of my patients, and that I wish them the best. It’s less obvious, I think, that sometimes the best plan is not what people want to hear, and doesn’t always feel safe to everyone. I base my decisions on a lot of information (that’s basically what psychiatry is–gathering as much information as possible). History, medical workup, mental status, collateral, medical files, team member assessments. I select the best plan of action from that. (Since I’m a resident physician at this time, all of my plans are discussed and approved by a staff physician. Eventually, that will not be the case.)

There is always bed pressure. I have to make sure that there is space in the hospital to treat the sickest should they arrive that night. Safety trumps that, but the bed situation is still a factor, and is steadily worsening–but that’s a separate problem.

Despite the large amount of info I gather, at some point I have to roll dice. Our risk assessment models do not have brilliant sensitivity or specificity, and there doesn’t seem to be a very reliable one out there. Exactly which information I’m supposed to consider is also a problem; there are lots of risk factors that co-relate, but retrospective data analysis has not proven definite causation to date.  Research in this area is ongoing, and perhaps we will find some good predictive tool. With the slow destigmatization of mental health, people are asking for resources in their communities and are reaching out for help. These things may improve. Clinical experience helps, but then again, it’s always the ones you never expect…

So. At some point, I might send someone home and disaster will unfold. And when I look back at the evidence I had, I very likely won’t find “the clue” to that. I do not have a crystal ball.

This has not yet happened to me. But really; I cannot provide a guarantee. Uncertainty will always reign over prognosis. Sometimes, we endure tragedy.

We are only human.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s