Diagnosis Made Easier, by James Morrison
Morrison’s book is a textbook for mental health clinicians. It is exactly what it says on the tin. It’s easy reading and could be read by a layperson with some knowledge of psychology. He lays out a number of important principles. I’ll just note a few that I find particularly useful.
In my own practice, I don’t often find diagnosis challenging in the sense of “What official diagnosis could this person possibly have?” (It has to do with where I work. Most of our clients were recently traumatized or court-mandated, so their issues tend to be pretty obvious - PTSD, depression, anxiety, life stress, etc.)
But the principles are still useful. For instance, I often have clients who match the criteria for multiple disorders. It is not at all uncommon for one person to fit the criteria for PTSD, generalized anxiety disorder, major depression or dysthymia, agoraphobia, panic disorder, and social anxiety. Which one is the most important? Is one causing all the rest? Which should I treat first? Which treatment is best for this person, given that they could in theory benefit from treatments for all the disorders? Is there one treatment that would cover several?
Also, sometimes the initial diagnosis is wrong or incomplete, and they need additional or different treatment. For instance, the client comes in with a clear-cut case of PTSD. But maybe they’re also non-neurotypical and have never been diagnosed. Or they were bulimic before and didn’t tell me about it. Etc.
Prioritization by safety hierarchy.
Since tests and investigation take time, you need to immediately check into some things, and leave others for later investigation. First, check for issues which are potentially life-threatening but treatable and likely to have a good outcome; last, check for issues which are not immediately life-threatening, hard to treat, and not likely to have a good outcome. In the middle, look for things in between. For instance:
1. Suicidality, major depression, bipolar disorder, medical illness. All potentially very dangerous, all potentially very treatable.
2. Panic disorder, OCD. (Very treatable but not immediately dangerous.)
Substance abuse without suicidality. (Potentially very dangerous, but difficult to treat; practically speaking, you can put a hold on the person who is threatening suicide, but you can't put a hold on a heroin addict just because you worry that they might accidentally OD.)
4. Personality disorders, dementia (if the person has caretaker and isn’t in danger of wandering into traffic or some such). Not easily treatable; may be life-threatening (Alzheimer’s) but immediate intervention won’t change that.
History beats current appearance
If you’re talking to someone who was raped last month, they will usually appear to have major depression, generalized anxiety, etc. Find out if they ever had any of those symptoms before.
Someone with a history of depression or anxiety is a better candidate for a psychiatric referral than someone who was fine pre-trauma. If someone was depressed before, treating their PTSD may not cure their depression, so they should be getting a med consult ASAP. (They are also much more likely to be open to the possibility of taking medication.)
If they have no previous history of depression and are not suicidal, sending them to a psychiatrist is not a priority; by the time they found an anti-depressant that worked for them, their depression may have resolved anyway because it wasn't true depression; it was a depressed mood caused by trauma. Treat the PTSD, and the depression goes away. Also, these people frequently don't want to take meds. (In those cases, I bring it up again if their PTSD seems to be resolving but they have persistent symptoms of some kind that are not improving with talk therapy. I point out that they may just need a little extra help in this one area, since we've tried our best with non-medical means and, unlike their other issues, it's not going away.)
And along similar lines:
Be cautious about diagnosing mental illnesses in people actively abusing substances when they do not have a known previous history of mental illness. Especially if they appear to be psychotic.
Maybe they’re schizophrenic. Maybe they’re bipolar. Maybe they’re depressed. But it is really hard to tell if they’re also snorting coke and drinking a bottle of whiskey every day. Very likely they do have some mental illness, but which one? Defer diagnosis and refer for specialized treatment (unless you’re the specialist.)
Get a family history
This is useful in so many ways. For instance, if you have a client balking at medication, knowing that their father was a drug addict may open up a conversation about exactly what their concern is - maybe they don’t know that SSRIs are not addictive. On a more obvious note, if there’s a long family history of depression, that makes depression a more likely diagnosis. Many mental illnesses run in families. (Typically, both genetic and learned factors are in play.)
Look for horses before you look for zebras
The teenage girl who seems anxious and pulls at her hair might have trichotillomania (the compulsion to pull out her hair.) But I’d check for anxiety, depression, trauma, neuro-atypicality, or even OCD first. (I'd start by asking her what the hair-pulling is about.)
Keep an open mind
Don’t just assume that previous diagnoses are correct. That includes your previous diagnosis. That also goes for family history. The “alcoholic grandfather” may have been self-medicating something.
Consider medical causes, including environmental ones
I once correctly diagnosed a client’s “panic attacks” as caffeine withdrawal. Morrison once thought he was suffering from a panic attack (he was in combat at the time, so very understandable) but it was actually an electrolyte imbalance - he’d forgotten to take his salt tablets. Once he took them and drank some water, his symptoms vanished.
Clients with chest pains need to see a medical doctor to rule out heart problems if they haven't already, even when you’re 100% sure it’s panic/anxiety. Depressed, low-energy, overweight clients and anxious, skinny, eye-bulging ones should get their thyroid checked. If there is any reason to think there may be a medical issue going on, send them to a doctor.
Always about ask medications a patient is taking, and ask again if they present with new symptoms with no clear cause. Specifically include OTC meds, herbal meds and supplements, and recreational drugs. Also ask if they’ve recently given something up.
Similarly, ask if they’ve been recently diagnosed with a medical condition. Many medical drugs and some medical conditions can have psychiatric side effects.
Clients presenting with a completely new condition in middle age with no clear cause (ie, not PTSD from a recent trauma) should be strongly considered for a medical cause. Mental illnesses (again, other than PTSD) tend to have a history, and middle-aged people are more likely to be physically sick than young people.
"Undiagnosed" is a perfectly legitimate label to use when you don't know what's going on, or don't have enough info to go on. Don't be embarrassed to use it.
The rest of the book is about diagnosing specific illnesses, such as “How to diagnose OCD.” It’s quite helpful (especially the diagnosis trees) but I thought the general principles above were more interesting for a general audience.
Diagnosis Made Easier, Second Edition: Principles and Techniques for Mental Health Clinicians
Crossposted to http://rachelmanija.dreamwidth.org/1232936.html. Comment here or there.
In my own practice, I don’t often find diagnosis challenging in the sense of “What official diagnosis could this person possibly have?” (It has to do with where I work. Most of our clients were recently traumatized or court-mandated, so their issues tend to be pretty obvious - PTSD, depression, anxiety, life stress, etc.)
But the principles are still useful. For instance, I often have clients who match the criteria for multiple disorders. It is not at all uncommon for one person to fit the criteria for PTSD, generalized anxiety disorder, major depression or dysthymia, agoraphobia, panic disorder, and social anxiety. Which one is the most important? Is one causing all the rest? Which should I treat first? Which treatment is best for this person, given that they could in theory benefit from treatments for all the disorders? Is there one treatment that would cover several?
Also, sometimes the initial diagnosis is wrong or incomplete, and they need additional or different treatment. For instance, the client comes in with a clear-cut case of PTSD. But maybe they’re also non-neurotypical and have never been diagnosed. Or they were bulimic before and didn’t tell me about it. Etc.
Prioritization by safety hierarchy.
Since tests and investigation take time, you need to immediately check into some things, and leave others for later investigation. First, check for issues which are potentially life-threatening but treatable and likely to have a good outcome; last, check for issues which are not immediately life-threatening, hard to treat, and not likely to have a good outcome. In the middle, look for things in between. For instance:
1. Suicidality, major depression, bipolar disorder, medical illness. All potentially very dangerous, all potentially very treatable.
2. Panic disorder, OCD. (Very treatable but not immediately dangerous.)
Substance abuse without suicidality. (Potentially very dangerous, but difficult to treat; practically speaking, you can put a hold on the person who is threatening suicide, but you can't put a hold on a heroin addict just because you worry that they might accidentally OD.)
4. Personality disorders, dementia (if the person has caretaker and isn’t in danger of wandering into traffic or some such). Not easily treatable; may be life-threatening (Alzheimer’s) but immediate intervention won’t change that.
History beats current appearance
If you’re talking to someone who was raped last month, they will usually appear to have major depression, generalized anxiety, etc. Find out if they ever had any of those symptoms before.
Someone with a history of depression or anxiety is a better candidate for a psychiatric referral than someone who was fine pre-trauma. If someone was depressed before, treating their PTSD may not cure their depression, so they should be getting a med consult ASAP. (They are also much more likely to be open to the possibility of taking medication.)
If they have no previous history of depression and are not suicidal, sending them to a psychiatrist is not a priority; by the time they found an anti-depressant that worked for them, their depression may have resolved anyway because it wasn't true depression; it was a depressed mood caused by trauma. Treat the PTSD, and the depression goes away. Also, these people frequently don't want to take meds. (In those cases, I bring it up again if their PTSD seems to be resolving but they have persistent symptoms of some kind that are not improving with talk therapy. I point out that they may just need a little extra help in this one area, since we've tried our best with non-medical means and, unlike their other issues, it's not going away.)
And along similar lines:
Be cautious about diagnosing mental illnesses in people actively abusing substances when they do not have a known previous history of mental illness. Especially if they appear to be psychotic.
Maybe they’re schizophrenic. Maybe they’re bipolar. Maybe they’re depressed. But it is really hard to tell if they’re also snorting coke and drinking a bottle of whiskey every day. Very likely they do have some mental illness, but which one? Defer diagnosis and refer for specialized treatment (unless you’re the specialist.)
Get a family history
This is useful in so many ways. For instance, if you have a client balking at medication, knowing that their father was a drug addict may open up a conversation about exactly what their concern is - maybe they don’t know that SSRIs are not addictive. On a more obvious note, if there’s a long family history of depression, that makes depression a more likely diagnosis. Many mental illnesses run in families. (Typically, both genetic and learned factors are in play.)
Look for horses before you look for zebras
The teenage girl who seems anxious and pulls at her hair might have trichotillomania (the compulsion to pull out her hair.) But I’d check for anxiety, depression, trauma, neuro-atypicality, or even OCD first. (I'd start by asking her what the hair-pulling is about.)
Keep an open mind
Don’t just assume that previous diagnoses are correct. That includes your previous diagnosis. That also goes for family history. The “alcoholic grandfather” may have been self-medicating something.
Consider medical causes, including environmental ones
I once correctly diagnosed a client’s “panic attacks” as caffeine withdrawal. Morrison once thought he was suffering from a panic attack (he was in combat at the time, so very understandable) but it was actually an electrolyte imbalance - he’d forgotten to take his salt tablets. Once he took them and drank some water, his symptoms vanished.
Clients with chest pains need to see a medical doctor to rule out heart problems if they haven't already, even when you’re 100% sure it’s panic/anxiety. Depressed, low-energy, overweight clients and anxious, skinny, eye-bulging ones should get their thyroid checked. If there is any reason to think there may be a medical issue going on, send them to a doctor.
Always about ask medications a patient is taking, and ask again if they present with new symptoms with no clear cause. Specifically include OTC meds, herbal meds and supplements, and recreational drugs. Also ask if they’ve recently given something up.
Similarly, ask if they’ve been recently diagnosed with a medical condition. Many medical drugs and some medical conditions can have psychiatric side effects.
Clients presenting with a completely new condition in middle age with no clear cause (ie, not PTSD from a recent trauma) should be strongly considered for a medical cause. Mental illnesses (again, other than PTSD) tend to have a history, and middle-aged people are more likely to be physically sick than young people.
"Undiagnosed" is a perfectly legitimate label to use when you don't know what's going on, or don't have enough info to go on. Don't be embarrassed to use it.
The rest of the book is about diagnosing specific illnesses, such as “How to diagnose OCD.” It’s quite helpful (especially the diagnosis trees) but I thought the general principles above were more interesting for a general audience.
Diagnosis Made Easier, Second Edition: Principles and Techniques for Mental Health Clinicians
Crossposted to http://rachelmanija.dreamwidth.org/1232936.html. Comment here or there.