Saturday, October 29, 2016

More Than 9 Questions About Sleep

I was just thinking about the PHQ-9 and it widespread use in managed care as a metric for depression.  The idea that 9 questions are all that is needed strike most psychiatrists as a gross oversimplification.  There has been plenty of debate over the years about the diagnostic criteria and waht should be included.  Kendler recently wrote an excellent paper on the fact that the diagnostic criteria as they stand in the DSM really indexes disorders rather than diagnosing them.  Some recent blog posts have looked at real patients and what is happening with them when they appear to have an elevated PHQ-9 score but are not depressed.  Past markers of psychiatric disorders like the dexamethasone suppression test had had to withstand more rigorous testing than the PHQ-9.  And lastly, the literature to support it seems to reflect the literature that justified managed care - a business concept with no basis in science or medicine.

And then I had the thought: "During an evaluation I ask more than 9 questions about sleep on the average."  By comparison the PHQ-9 has one question.  That question is:

3. Trouble falling or staying asleep, or sleeping too much?  
[Not at all] [Several Days] [More than half the days] [Nearly every day]

There are more complex sleep questionairres.  The Pittsburgh Sleep Quality Index (PSQI) and Functional Outcomes of Sleep (FOSQ, FOSQ-10) are good examples.  I thought I would tabulate my questions here.  Sleep disturbance can be a primary disorder independent of any psychiatric problem.  Sleep is also comorbid with many if not most psychiatric problems.  Most people do not recover from a priamry psychiatric disorder as long as their sleep is disturbed.  Sleep disorders can antedate the onset of mood and anxiety disorders by years and for that reason I think it is important to determine if the sleep disorder is primary rather than part of the mood or anxiety disorder.  That cannot be determined by a brief cross sectional look that considers all current symptoms as part of a mood disorder.  So during my standard evaluation I ask people the following questions about their sleep not necessarily in the following order:

When you were a kid in middle school or high school did you have trouble sleeping?
Did you have nightmares back then?
Did they occur early in the night or later in the early morning hours?
Did you sleepwalk?
Did anyone ever tell you that you had sleep terrors?
Did your sleep problems from childhood ever resolve - have you ever slept normally since then?

Do you work at night?
Do you do shift work where the work time changes?
Do you currently have sleep problems?
Do they occur when you try to fall asleep?
Tell me your experience of trying to fall asleep - what gets in the way?
Any idea how long it takes you to fall asleep?
Have you had sleepless night where you could not sleep at all?
Do you wake up off and on all night long?
Do you wake up early in the morning - like 4 or 5 AM and find that you can't fall asleep?
Is your sleep restorative - do you feel rested the next day?
Do you snore?
Have you ever had a sleep study?
Have you been diagnosed with a sleep disorder?
Do you take alcohol or any medication to help you fall asleep?
Do you take in many caffeinated beverages during the day as coffee, soda, tea, or energy drinks?

That is about 20 fairly basic questions about sleep.  It is a framework that requires elaboration.  Just the issue of sleep studies these days can lead to details about parasomnias, related surgeries, sleep disordered breathing diagnoses, restless leg syndrome, and all of the associated treatments.  For the pupose of this post that is about 18 questions or twice as many as the total on the PHQ-9. It should be apparent that severe sleep problems can lead to a score considered in the depressed range on the PHQ-9 by adding up the scores of questions 1, 3, 4, 7 and 8.  Critical distinctions need to be made between sleep problems, anxiety disorders, mood disorders, addictions, and also the excessive use of an adult attention deficit-hyperactivity disorder diagnosis.

This brief exploration should point out he problems with a screening versus a diagnostic evaluation.  If you are given a PHQ-9 or GAD-7 (for anxiety) in your primary care clinic that score alone is insufficient as a basis for a treatment plan for depression.  A person repeating the questions as confirmation does not constitute a diagnostic evaluation.  By itself it does not mean that an antidepressant prescription is indicated.

Ask that person how they know that you have depression and not a sleep disorder or something else.

George Dawson, MD, DFAPA


  1. The PHQ-9 is invalid on the face of it since it in no way addresses the important questions of pervasiveness, persistence, difference from a given patient's normal everyday moodiness, sleep, appetite and energy issues (and how long this difference has been going on), or even whether the patient has all of the symptoms at the same time.

    It is a screening test, not a diagnostic test, and as such was actually designed to have a lot of false positives. When I worked at the VA, they wanted the psychiatrists to do them, and I flat out refused. The test was meant to triage patients for further psychiatric evaluation, and since they were already getting one when they saw me, it was absolutely superfluous.

    Psychiatrists who agree to use this test need to have their own heads examined.

  2. PHQ-9---psychiatry's answer to Pain Management's Ouchie Scale for Kids.