Friday, July 24, 2009

Can Black Box Warnings Kill?

I'm going to write about a story I saw on-line about a depressed mother who poisoned her small child. It's a terribly tragic story, and please keep in mind that I only know what I read in the article Here, and I've never examined anyone involved. The question being asked at the trial is that of whether the mother, who was depressed, was legally sane and knew it was wrong to kill her child, and that's not what I'm going to write about. I didn't pick a graphic to go with this blog post, because I couldn't think of any photo that would be appropriate to such an angst-ridden topic.

I'm pulling a few sentences from the newspaper article to use as a springboard for discussion:

They said Sparrow told a nurse practitioner she was considering using sedatives to kill herself, her daughter and her dog, but that medical professional did not contact the authorities or otherwise try to get Sparrow committed to a psychiatric hospital.

After hearing Sparrow had just stopped taking the antidepressant Prozac for fear it was causing the suicidal thoughts, the nurse practitioner let her go home with the instruction to come back if she didn't feel better...

I was struck by two things in the recounting of the story as I read it: that both the patient and the nurse practitioner thought her suicidal thoughts came from the Prozac (and both, perhaps, trusted they would stop with the cessation of the medicine--- obviously I don't know that's what they thought, but it's implied in this particular recounting of the story), and that a homicidal mother was apparently allowed to leave a clinic without being evaluated by a psychiatrist, I think. So my comments are general, because I don't trust a press account to be all-inclusive, and perhaps things transpired that didn't make it in to print.

When Prozac first came on the market, there were some concerns that it made people suicidal, and these concerns were dismissed. With years (oh, more than a decade) researchers revisited this idea and concluded that people under the age of 26 have a low incidence (1-2%) of violent thoughts caused by anti-depressants, and so we have the Black Box Warning about such thoughts. Does all the publicity about how the possibility of suicidal thoughts can arise from the medications narrow peoples' thinking? If we think a medication has caused a suicidal idea, does this prevent people from exploring other options? Perhaps the medication isn't working, or perhaps the depression has gotten worse and has broken through. Perhaps something else has transpired that increases risk. And if the medication is the culprit, what do we know about how long one has to be off it before such violent thoughts stop and the risk is gone? I think the answer is that we don't know.

I don't know if the woman described above saw a doctor the day she was in the clinic, or what exactly she said to the nurse practitioner. I don't know if the outcome would have been any different if she'd been committed to a psychiatric facility. What I do know is that when any story has a tragic ending, it's hard to wonder if more couldn't have been done.

We pass so-called scope-of-practice laws--- should psychologists prescribe? Should nurse-practitioners practice essentially independently? The fuss goes into the legislative battles before-the-fact, one fought primarily by legislators and lobbyists, not clinicians. We don't generally look backwards and ask if poor outcomes are more more likely to occur in settings where we've dropped our standards and we don't seem to ever ask if we should revoke those decisions. I'm not saying we should--- but perhaps we should ask more questions.


Anonymous said...

Hi Dinah,

Warning - an angry rant is coming.

Why, why, why do you and psychiatrists refuse to believe that antidepressants can cause suicidal and homicidal ideation and blame it on the person's label?

If someone developed suicidal and homocidal ideation from taking a regular med and didn't have a psych label, they would be taken off the drug in a heartbeat. The person wouldn't be accused of getting ideas from black box warnings.

Frankly, to say that someone is influenced by black box warnings comes across as condescending even though again, I am sure that wasn't your intention. As one who also suffered suicidal ideation from Prozac, it sounds like this woman's perceptions were correct but sadly, cold turkeying off the med made the problem worse.

By the way, I am stunned you didn't realize that cold turkeying Prozac might have worsened the suicidal ideation.

So perhaps an evaluation by a psychiatrist would have saved her life. Then again, she could have had a similar situation like mine in which my meds were doubled by a former psychiatrist which severely worsened my condition. It goes back to the mentality of blaming the person's label for the condition and not the meds.

Finally, as an FYI, I don't always feel the med is the reason for suicidal or homicidal ideation. But while I agree with you that there is alot we don't know, it seems like in this case, it was definitely an issue.


Dinah said...

Anon: I do believe that SSRI's can cause suicidal thoughts, that wasn't the question.

The issue was that the patient and nurse practitioner assumed this to be true without exploring how it might play out and if there were other options. You sound like you're set in believing it was the medications, and not worsening depression or, according to the state's experts in the article, what could have been a plotted=out revenge killing. I believe the mother told the nurse she planned to poison the family, had stopped the prozac, and acted on the idea weeks later.

Cause is not the salient issue here-- when someone talks about poisoning their family, it's generally a reason to put them into the hospital for further evaluation and to ascertain safety. Even it was caused by a medicine, and even if it will wear off in a fantasy 24 hours, everyone needs to be safe until the thoughts/intent ends or there is some mechanism in place to be sure everyone is safe.

I don't know why this mother was talking to the nurse about poisoning herself, her child and her dog, but everyone involved may have lulled into thinking it was "Just the medicine" (my quotes for emphasis) safe for her to go home because of the thought that the symptom was caused by the medication, and even if it was the medicine, well, it obviously was not safe (per monday morning quarterbacking).

Medicine may have all kinds of side effects, but other possibilities should also be considered when a symptom develops or worsens.

Anonymous said...


I am really trying to understand your point of view but I continue to be stunned by what you write.

If someone waits to see how a side effect from a psych med plays out, it might be too late.

Cold turkey/quick tapering side effects can last for weeks and not days. I

With all due respect, you and your colleagues needs to broaden your thinking about how long withdrawal symptoms can last.

I am not disagreeing that she should have been hospitalized. But because I feel psychiatrists are clueless about withdrawal issues, I have doubts that the outcome would have been different.

Now, I will admit that the article left out a key piece of information which is what her history was prior to being placed on meds. If she still had these evil dark thoughts, then I would agree there is more going on.

Also, the key is if her family members and friends express shock at her crime assuming they aren't in denial and didn't miss key warning signs.

But even if she had a history, the med could have pushed her over the top although I wouldn't look at it as the main culprit.

TheCrazyMusicLady said...

I'm confused because in your first comment, you say that assuming someone is influenced by the black box warning is condescending, but then go on appearing to state that the medication must have been the culprit.

What I think Dinah is trying to say is that from the position of the NP and other practitioners in this case, does the presence of the black box warning create a knee jerk reaction to SI/HI instead of leading the practitioner to examine other possible causes. Working where I work (inpatient C&A psych), if we assumed that every occurrence of SI/HI that came when a child was on SSRIs, we would have children in the hospital for months trying to stabilize them.

Also, Dinah, I'm asking you this because this is what I understand from the psychiatrists I work with. SSRI's, if they are to cause a worsening of or appearance of SI/HI, it generally happens within the first week to a month of taking the medication. Not being a psychiatrist myself, I only know what I learn from my colleagues, it just seems that if this is the case, then shouldn't the idea of this being caused by the 'black box' effects be disproven?

Just some notes :)

Ally said...


I am not going to get into the controversy about this particular case, just wanted to get your take on something:

When does a decompensation after SSRI use lead to revising the dx to bipolar? How does a doctor know whether to try a new SSRI or get off that track entirely and move to mood stabilizers, etc? Just asking because this was my experience, change in dx and treatment plan.


Annie said...


I have enjoyed your blog and podcasts for the last year.
How disappointing then, that you chose to make an issue about the "nurse practitioner" in the case, suggesting, not so subtly, that she was negligent and then segueing into the tired argument of whether anyone else, other than a physician, can give good care and make thoughtful medication decisions.
As everyone, including yourself, has said, you do not know the details of the case but are making an argument against the practitioner based on a small news article. Nurse practitioners have been caring for patients for a long time now and many are in specialty areas, like acute care, cardiac care and psychiatry.
Please remember that there are many nurse practitioners and physician assistants who take very good care of their patients and there are probably an equal amount of physicians who don't. Nurse practitioners have documented evidence that their outcomes are often better than their physician colleagues.This discussion should be about the decisions made, not used as a rant against NP's, who are medically trained and licensed to prescribe nor against psychologists, who aren't medically trained and want prescribing power.If this was one of your psychiatrist colleagues, would you give him or her the benefit of the doubt, knowing how little you are informed about the specifics of the case? Full disclosure - I obviously am a psych NP.

Midwife with a Knife said...

I agree that the main point is that the black box coulda killed someone, in this case.

Also, I think that saying that nurse practitioners are the equivalent to physicians is misleading and depreciates the value of the extensive hours and challenges of training of an MD.

Anonymous said...

It is interesting to me that if a patient says a medication is helping them that psychiatrists take that at face value and believe it's true. Conversely, if a patient says the medication is increasing suicidality that there's a need to "explore" that.

Anonymous said...


Anonymous said...

Isn't this a lot like...

"what comes first, the chicken or the egg?"

Especially with these meds that are commonly given to people who could easily be having suicidal or violent thoughts? or on their way to having them?

There really is no way for anyone to know if the thoughts are there because of the medicine or because their condition is simply getting worse or the medicine is not working... just like the chicken in the egg... :)

Dinah said...

Personally, I think he's easy to Monday morning quarterback when a child dies.

If we take the story at face value, the nurse practitioner believed that the prozac was causing the violent doesn't really matter What caused the violent thoughts, if they are so formed as to entail a plan to kill two people and a dog, the patient needs to be somewhere safe. If in fact, the treating provider let this patient go without a second opinion (and I don't know that from the article), my guess is that when the child was killed, he/she may have wished she'd gotten another opinion. If the patient was homicidal because she was depressed (as opposed to SSRI induced) or if the ideas were due the meds, we can't assume that any pharm change (stopping ssri's or changing the medications) would be enough to prevent a horrible outcome soon enough. If I were the doc seeing this patient, I would have regretted not hospitalizing her or at least getting a second opinion. I've had bad outcomes, and I've second guessed myself, but they were unpredictable: the patient did not tell me they were going to commit a violent act.

I have no problem with taking a patient off a medication they believe is causing suicidal ideation (clearly the medicine was not helping!) A murdered child implies that something more was needed.

This story reminds me of a time I was called the the ER to consult on a patient who was found walking naked on a major highway. The evaluating social worker told me the plan was for discharge because the patient was already scheduled for an out patient evaluation in the morning at another facility. I was to ask why he believed that someone walking naked on a highway would live to make it to the morning.

I don't mean to be judgmental of this case, I don't know the details and I don't believe the press presents all the sides of every case. Sometimes stuff happens that mental health care workers could not foresee, and the journalist's presentation led us to believe this was foreseeable and that the person treating the homicidal woman should have detained the patient-- so using that as my maybe-tainted source, I'm left to monday morning quarterback.

It's not about nurse practitioners- a psychiatrist could have made the same mistake (to discharge) and any one who hears a story of a depressed mom wanting to kill a child should either admit the patient or a second opinion, or rally the support of family to watch the patient. I let a post-partum suicidal patient go home once-- her husband was in the session, he was instructed to watch her, she was given a fast acting medicine, and they were both instructed to return in the morning (which they did).

My main point was that the publicity of the black box warning MIGHT narrow our thinking and keep us from entertaining a range of possible causes and treatments to the issue of suicidal ideation.

I have to say, people talk about suicidal thoughts regularly, and it's a rough call to know when they might act on them. Clearly, we can't hospitalize everyone who talks of dark thoughts, we can't even get them all second opinions. We do our best and hope our careers never see a tragedy such as this. But we all know, no matter how wonderful we are as psychiatrists, that there is an element of uncertainty, and bad outcomes happen to the patients of the best of docs.

Franzi said...

Thanks for not jumping to conclusions. I'm so sick of docs who take every opportunity to rake an NP over the coals while denying every irresponsible, neglectful, or ignorant thing doctors do.

I'm not a nurse or related to or close friends with a nurse for the record.

Sunny CA said...

I look at the use of a nurse practitioner and the failure to provide hospital care as have the same root cause: a desire of somebody to save money on care. I once belonged to a medical plan which never provided me with an physician. I saw a n.p. for gyn, a non-physician for eye checkup both of which I had previously had used a physician for. If I was sick I saw a n.p. This was in 1980, so not something new. My guess is that this medical group would also use an n.p. instead of a psychiatrist and that they would tell the n.p.'s to all ration hospital care. When the insurer is also the gatekeeper to the services received it seems likely that care would diminish. They likely have a policy that if the person isn't actively trying to kill himself or actively trying to kill someone else, then it's a wait and see situation because it saves money.

Anonymous said...

I've been on three different antidepressants: Effexor, Wellbutrin, and Lexapro.

The Effexor made me feel "weird" and I would get upset about trivial things, and I recognized my reactions were out of proportion. So the doctor switched me to Wellbutrin.

I took it for a full year and felt totally normal until the end, when I started getting thoughts "I should kill myself". What's strange is there was no emotion connected to these thoughts, they were just sort of matter of fact. I didn't actually FEEL like killing myself, it's just that those were the WORDS in the thoughts. So I went to the doctor immediately and was switched to Lexapro.

Lexapro was wonderful, I took it for two years, but finally got off of it after starting therapy and deciding I wanted to feel my feelings, and the Lexapro was preventing that by making me a little emotionally numb.