Imagine that you got upset. Is it very remarkable that I can “diagnose” that you are upset? After all, you are clearly upset. What expert thing did I accomplish by agreeing with you that you were upset?
Imagine that you are angry. Is it very remarkable that I can “diagnose” that you are angry? After all, you are clearly angry. Have I added anything meaningful by saying “I diagnose that you are angry” instead of “You seem angry”?
“You look upset” is the simple, truthful thing to say, and “I diagnose that you look upset” is a piece of self-serving chicanery. By adopting that circumlocution, I’ve tried to turn an ordinary observation into a pseudo-scientific marvel. If this is the way I’m operating, I dearly hope that you won’t notice my little game.
By contrast, let’s say that you explain to me that you’ve been having hallucinations. You describe the look of your hallucination, and you also describe to me your recent history, other physical symptoms, and so on. Taking that information together, I have a strong hunch that you’re suffering from early Parkinson’s. I then run tests to confirm or disconfirm my hypothesis. I didn’t “diagnose” your hallucination—you handed me that. I diagnosed your Parkinson’s.
We seem to have a lot of trouble understanding this difference: the difference between “diagnosing a symptom” and “diagnosing a cause.” The second is what medicine legitimately does. The first is what the mental health establishment illegitimately does. It is not real diagnosis for me to “diagnose you with an anxiety disorder” because you told me you were anxious. This is chicanery and not diagnosis.
You don’t diagnose symptoms. You diagnose causes. To diagnose a symptom is only to say, “Yes, I agree, you have a rash.” Everyone who looks at you knows that you have a rash! What we want to know is what sort of rash it is. What’s causing it? You observe the tumor and diagnose the cancer. You observe the bump and diagnose the concussion. You observe the fever and diagnose the influenza. You don’t observe the anxiety and diagnose the anxiety. That is wrong.
You observe a symptom, you interpret a symptom, and you make use of a symptom as part of your efforts at diagnosis. But the symptom isn’t the diagnosis. You observe a symptom and then diagnose a cause. You don’t observe anxiety and then diagnose anxiety. It isn’t okay to call this “diagnosing.” It isn’t okay to turn a report of anxiety into “an anxiety disorder” just by saying so. Yet this is what is done all the time nowadays.
Here, for example, are some of the questions whose positive answer will get you an “anxiety disorder” diagnosis:
“Are you feeling keyed up or on edge?”
“Do you have feelings of panic, fear, or uneasiness?”
“Are you constantly worrying about small or large concerns?”
“Are you constantly tense?”
“Does your anxiety interfere with your work, school, or family responsibilities?”
“Are you plagued by fears that you know are irrational, but can’t shake?”
“Do you avoid everyday situations or activities because they cause you anxiety?”
“Do you watch for signs of danger?”
If you answer yes to these questions, you are acknowledging in these different-but-same ways that you are feeling anxious. But what you get from the mental health establishment is not, “Yes, you are clearly feeling anxious. Let’s see if we can figure out why.” What you get is a “diagnosis” of an “anxiety disorder.” In our current system, you appear to have “symptoms” of an “anxiety disorder.” You come in looking anxious, acting anxious, and saying that you are anxious. What sort of diagnostic acumen does it take for me to say, “You’re anxious”?
Let’s look at a hypothetical fellow. Jim has looked a certain way his whole life: he has been recognizably himself for as long as he or anyone can remember. If you are his mate or his friend, you are pretty much certain how he’ll react if, say, you offer him a second beer, ask him if he wants to climb a mountain, ask him to play a board game, and so on. You know his likes and dislikes, his habits, his characteristic expressions, pretty much everything.
Jim goes away for a week and comes back different. This is a classic plot in fiction, where a character goes off for many years, say to war, and when he comes back his wife is certain that he isn’t the same man but an impostor, even though he looks the same, has all the right memories and information, and can pass any test on being himself. However, his wife just knows that he isn’t the same man. Let’s say that Jim, who has never seemed particularly anxious previously, comes home highly anxious.
He looks anxious. He says he’s anxious. His anxiety is keeping him from sleeping. The question isn’t, “Is Jim anxious?” The question is, “Why is Jim suddenly anxious, so much so that he doesn’t even seem to be the same person?” To “diagnose” Jim with “an anxiety disorder” is child’s play: he is clearly anxious. That is no diagnosis at all. That’s like diagnosing Jim with “lump-itis” if he comes in with a baseball-sized lump on the side of his head. “Lump-itis” won’t do. Nor will “anxiety disorder.”
We want to know three things: Why is Jim anxious? What might we suggest to help Jim, given the particular source of his anxiety? And what can we suggest to help Jim whether or not we’re able to discern why he is anxious? Too many mental health providers skip the first two questions because Jim may not cooperate in their investigation, because those “causes” are too hard to discern, and/or because those “causes” are frankly unknowable. They go directly to the third. They say, “Let’s treat that anxiety!” They say, “Let’s treat that symptom!” The pills appear; talk of a certain sort commences. No real investigation of the cause begins because steps one and two are assiduously skipped.
Jim’s Week
Let’s imagine that the following ten things occurred in Jim’s life during the week that he was away:
1. On the flight home, his plane encountered engine trouble and had to be diverted to Amarillo.
2. He started an affair with a woman half his age and half his wife’s age, an affair that he would like to continue even though he feels guilty about betraying his wife.
3. He took certain street drugs for the first time.
4. He had a memory that he couldn’t shake about a test that he failed miserably in fourth grade and how his teacher humiliated him when she handed out the test results.
5. He experienced certain physical symptoms, including heart palpitations and sudden sweats.
6. He received an email and learned that he was about to be audited.
7. He had a nightmare in which he saw himself being drawn and quartered for his unpopular beliefs.
8. He received an email from his sister saying that she was unwilling to continue taking complete responsibility for caring for their demented mother.
9. He began wondering if misplacing his hotel key and his car keys were signs of his own early dementia.
10. He reread a portion of a novel he had once written, discovered that he still liked it, and began to wonder if he should resume writing it.
Whether or not any of these “caused” Jim’s sudden anxiety, doesn’t each feel suggestive and something like a potential clue? Wouldn’t you want to know these things if you were tasked with helping Jim reduce his experience of anxiety? Forgetting for a second about how you might actually discern which of these, if any, was causing Jim’s sudden onset of anxiety, don’t we suspect that even just getting them named and “on the table” might have some salubrious effect on Jim? Doesn’t all therapy that isn’t caught up in “diagnosing and treating mental disorders” rely on this central idea, that making the unknown known helps people reduce their experience of distress?
We’d certainly like to possess that information about Jim’s week. Let’s add another wrinkle. What if Jim reports that he believes that the anxiety is the result of him being followed all week by a certain suspicious stranger—a man Jim is sure was there but who in fact wasn’t. In this scenario, Jim may be suffering from an actual organic syndrome—and a feature of this organic syndrome may be that Jim will not believe you when you try to explain to him, if you are in a position to do so, that the stranger does not exist and could not have existed.
For example, patients with Anton’s syndrome, which can arise in blind individuals with cortical damage, may “see” exactly such strangers as Jim feels he is seeing—and they can’t be convinced that they are hallucinating. Oliver Sachs explained in Hallucinations, “A patient with Anton’s syndrome, if asked, will describe a stranger in the room by providing a fluent and confident, though entirely incorrect, description. No argument, no evidence, no appeal to reason or common sense is of the slightest use.”
This possibility should further highlight what investigating looks like. You can’t learn this vital information if the transaction plays itself out in the following way, as it almost certainly will between Jim and a chemical-oriented psychiatrist:
“So, Jim, you’re generally anxious?” says the psychiatrist, looking at the intake form.
“I am!”
“That means you have generalized anxiety disorder.”
“Okay!”
“As it happens, I have some yellow pills, orange pills, and blue pills to treat that disorder. Let me tell you about them.”
“Can’t wait to hear!”
It should be perfectly clear that “diagnosing the symptom” (“You’ve got anxiety!”) and then “treating the symptom” (“Here’s a pill!”) is simply the path of least resistance. We can see why it is so tempting to engage in this shortcut and this illegitimate process since it appears well nigh impossible to know whether Jim is anxious because his anxious nature, dormant “forever,” suddenly kicked in; whether a single idle memory, say of that fourth grade humiliation, caused “all this emotional fuss”; whether the affair, the audit, or the near plane crash provoked this new anxiety; and so on. Rather than admit that he doesn’t know what is causing the anxiety, probably can’t know, and doesn’t really care one way or the other, a chemical-oriented psychiatrist simply proceeds to “diagnose and treat the symptom.”
We understand the temptation. There are no tests to connect these possible “causes” in any direct or indirect way to Jim’s sudden onslaught of anxiety. For the most part, we simply can’t know. Nor can Jim, for that matter. Maybe he reads over his list and exclaims, “It’s the audit! That’s what’s doing it!” Do we trust Jim’s judgment on this score? Do we believe that Jim is accurately ascribing the right cause from among all these plausible causes? Do we believe that a certain feeling, insight, or self-report on Jim’s part amounts to an “accurate diagnosis” of the source of his anxiety?
Not if Jim is the tricky, complicated, evasive creature he surely is—namely, not if Jim is a human being. Can we really trust Jim, especially once we discover that he doesn’t want to talk about how the new affair may be affecting him? Do we want to “diagnose” Jim with “audit-induced anxiety” because he’s cherry picked the audit off his list? No. We know better than to take at face value the often self-serving explanations that human beings provide. We can’t rely on Jim; we can’t “test” Jim. And so, then what? Do we throw Jim to the chemical-oriented psychiatrist? No. Rather, we say to Jim, “You know, I wonder . . .”
This “wondering” might sound like the following:
“Jim, you’ve picked the audit as the main source of your anxiety. But what about the affair you’re having and what about keeping that affair a secret from your wife? Isn’t that likely implicated? Don’t you think you might be suffering from a guilty conscience?”
“I don’t feel guilty.”
“But you said earlier that you did feel guilty.”
“I do and I don’t. The bottom line is, I don’t want to end the affair, and I don’t want my wife to know. I want to get rid of the anxiety, not end the affair.”
“Oh. That is so interesting.”
This is a perfectly clear, characteristic, and plausible exchange. Jim’s part of the exchange can be translated as, “I want relief from the symptom, and I do not want to make any fundamental change in the way I’m operating. If you happen to have a pill handy, that would be lovely.” Jim may well pull for that pill! And what frustrated helper wouldn’t want to hand Jim the pill bottle and say, “Have it your way. This pill has strong effects that may quell your anxiety. I wash my hands of you and your nonsense. Let’s collude in acting like the anxiety is a ‘thing’ that blew in through the open window and that we have ‘medicine’ for it. Fine. Let’s play that game.”
However, our human experience specialist, or any psychotherapist who doesn’t feel obliged to “diagnose and treat symptom pictures,” doesn’t have to pull out the pill response. Our specialist can continue, one human being to another, “You want the affair, you want the lying, the hiding, the cheating, and all of that, and you don’t want the anxiety that may come from the lying, the hiding, the cheating, and so on? Have I got that right? Tell me, Jim—does that make any sense to you?”
You investigate, you suggest, you hypothesize . . . and you tell the truth. There is a world of difference between “diagnosing and treating” and investigating, suggesting, hypothesizing, and telling the truth. When psychotherapists investigate, suggest, hypothesize, and tell the truth, they are helping; when they “diagnose the mental disorder of generalized anxiety disorder,” they are playing a game, illegitimately labeling, and creating a pseudo-patient who, like Jim, may prefer to be a pseudo-patient than deal with the turn his life has taken. When they do the former, they deserve a well-earned round of applause; when they “diagnose,” they ought not sleep well.
Consider the following. What if nothing unusual or provocative happened to Jim during his week away? Does that make Jim’s anxiety uncaused? Of course it doesn’t. It only means that we know even less about its source than if we had some obvious clues. In a certain sense that may even prove helpful, because without obvious clues, we can’t leap to connecting up Jim’s sudden anxiety with some too simple “cause.” We would naturally presume that there are reasons for his heightened anxiety, reasons that we may never come to know, and even more adamantly invite Jim to collaborate in our investigation.
Let’s say that Jim does collaborate and that he lands on that audit as the source of his anxiety. We may not completely believe him, but should we dispute him or ignore his formulation of the problem? No, we shouldn’t. After we’ve said what we had to say, for example, about his affair, we might then want to take Jim at his word and consider the possibility that the upcoming audit is indeed the primary source of his anxiety. Given that Jim has said so, we might take that as a working hypothesis. A working hypothesis is very different from a diagnosis. When a doctor says, “It might be this, it might be this, it might be this, or it might be this,” he is announcing his hypotheses. He hasn’t made a diagnosis yet. Nor should we as we begin our investigations. In medicine, you don’t diagnose until you can diagnose.
Will a moment come, in Jim’s case or in any other, when we can “make a diagnosis”? I think the answer is a clear no. What could such a diagnosis sound like? Audit-induced anxiety reaction coupled with denial-induced extramarital affair-itis? Sudden Personality Change Syndrome caused by a plane’s engine catching fire? Unleashed Primal Lust for Younger Woman Syndrome with overtones of guilt and pleasure? These human events can’t and shouldn’t be “diagnosed” as if they were illnesses. Let us stop looking to diagnosis as the Holy Grail. It isn’t. Right now it is only a mechanism for turning human experience into mental health establishment profits.
Not Diagnosing at All
For the many professionals looking for decent alternatives to the current mental health labeling system and our methods of diagnosing, I would say that the best alternative to our current ways of diagnosing is to not diagnose at all. Let us understand diagnosing for what it is: an inappropriate lifting of a term and an idea from enterprises where it makes sense, like repairing cars and repairing hearts, to one where it doesn’t. We will do much better as helpers without adding illegitimate “diagnosing” to our game plan.
A lawyer helps clients without diagnosing or labeling. An accountant helps clients without diagnosing or labeling. Our new human experience specialist can also help—as can any psychotherapist who rejects the pseudo-medical model—without diagnosing or labeling. If they are suffering from a medical illness, they should see a doctor, who has the job of diagnosing their Parkinson’s or their Anton’s syndrome. Diagnosing those real illnesses isn’t the job of any psychotherapist on earth. Nor should it be their job to “diagnose” made-up ones.
The end of diagnosing doesn’t imply the end of helping. It does imply the end of “treating,” another medical word and idea. It would actually promote rather than impede helping. In a postdiagnosis future, you could say, “You’re anxious. Let’s investigate why because maybe that investigation will help us. And if we can’t figure out why, that’s okay too. We can still try out some things that may help you feel less anxious. And, who knows, they may even help with what’s causing the anxiety. Okay? By the way, I will need your cooperation in this because I am not a doctor with tests and treatments and what-have-you. I am just a person like you. I’m willing to focus you a little, ask some pointed questions, be on your side when I feel I can be, and be ‘in this’ with you. But I need your help. Okay?”
This new human experience specialist would probably also have to say to Jim, “By the way, let’s talk about pills for a second.” That is, she would probably have to put on the table the fact that Jim can take certain chemicals-with-powerful-effects that might quell his anxiety. Maybe she’ll have at her disposal a newly minted brochure sanctioned by the government that she could hand Jim that would spell out the “role of chemicals in altering human experience.” At any rate, society might well demand that she have a pill chat with Jim and inform him of his right to take chemicals to deal with his anxiety. But let’s leave the details of that difficult discussion aside for the moment.
To return to the main thread, can we help Jim even if we can’t locate the “cause” of his anxiety? Can we legitimately decide to “treat” the “symptom” without identifying the “cause”? Of course we can. You can “treat a symptom” without “diagnosing a symptom.” This happens all the time in medicine. You call your doctor late in the afternoon and say, “I have a headache.” He says, “Take two aspirin and call me in the morning.” He has not diagnosed you, but he is treating your symptom. History tells him that you will probably wake up without the headache for any one of three reasons: that the aspirin worked, that some placebo effect worked, or that the headache, like many headaches, just went away.
It is perfectly plausible, sensible, and reasonable to sometimes “just treat symptoms.” We have plenty of things to recommend that might help a sad person, an anxious person, a person who drinks too much, and so on. Maybe a sad person would benefit from some sunlight, an anxious person from some “don’t sweat the small stuff” training, a problem drinker from AA. But if the headaches persist, you don’t just keep “treating the symptom”—you don’t just keep recommending aspirins. You say to the person across from you, “Please, please, please, let’s see if we can get at what’s going on here. Okay? Can we please do a little investigating?”
In real life, you can do both: you can “treat the symptom” by providing some time-tested tactics, and you can also “investigate what’s going on underneath.” Indeed, this amounts to best practice. And nowhere in this best practice was there a need for diagnosing to rear its ugly head. Yet many well-meaning mental health professionals retain a desire to diagnose because they genuinely believe in the “diagnosing and treating” model. A well-known therapist dropped me the following note in response to a column of mine in Psychology Today:
I appreciate your position and understand your concern about using the word “diagnosis,” but I think it is unfortunate that we have ceded this term entirely to medical practice. According to Merriam Webster’s dictionary, diagnosis is the “investigation or analysis of the cause or nature of a condition, situation, or problem.” By this definition, mental health disorder categories may not even qualify as diagnoses because by being “atheoretical” regarding cause, they offer little in terms of meaningful explanation that lends itself to helpful courses of action.
Psychological formulation, on the other hand, fits nicely with the idea of diagnosis as a process by which we come to identify and understand the problems clients present to their therapists. I’d go so far as to say that based on this definition, client and counselor could even diagnose a problem together—something quite different from the usual presumption that diagnosis is limited to the notion of expert therapists independently deciding what is wrong without client involvement. My point is that diagnosis need not be viewed in the very restricted sense it has been for so long in the helping professions.
I understand this desire, but this would still amount to a misuse of the word “diagnosis,” and it would continue our current pattern of abusing both language and human beings. You discern causes, if you can; you don’t co-create them. A human experience specialist and her client can certainly co-create a plan for managing anxiety, co-create an agreement about what the client will or won’t try, and so on. All of that can be co-created. But you can’t co-create a diagnosis. That’s Monty Python territory.
The challenge for any contemporary psychotherapist who wants to retain an ability to “diagnose and treat” is simple to describe: give me an example of your updated diagnostic system. Tell me how you would test to confirm your diagnoses and how you would distinguish one cause or source of a problem from another cause or source of a problem. Give me your taxonomy—your naming system and your rationale for using it—and let’s hold it up to scrutiny. If you want to continue diagnosing, put up the names of your “mental disorders” and let’s look them over. And don’t forget to clearly indicate what you are counting as causes! If you don’t take causes into account, you still aren’t really diagnosing. You are merely cataloguing.
I think that we will discover, if we are truthful and if we are acting in good faith, that it is impossible to retain the idea of “diagnosing” when it comes to human experiences. We should stop “diagnosing symptoms” right now, as that is a completely illegitimate enterprise that is annually adding millions of people, many of them children, to the rolls of the “mentally disordered.” This should stop today. But we should also let go of the idea that “diagnosing and treating” makes any sense in the context of human experience. It is simple: we have adopted the wrong model. It is past time to discard it.
As to whether there is perhaps some way to retain the idea of “diagnosing,” let those who want it retained describe what their taxonomy might look like and let us see if we believe them. I don’t think we will believe them, because it is folly, and always will be folly, to “diagnose the human condition” when we have no way of knowing what counts as cause and effect in human affairs. Are we to “diagnose” personality differences, changed circumstances, stray and odd thoughts, and every single human thing, from war breaking out to a month of cloudy days? Such an enterprise makes no sense.
We do not know what caused Jim to become anxious, and while we can investigate his situation with him, we can’t arrive at the sorts of conclusions that in medicine are called “diagnoses.” To announce that we can arrive at such conclusions or that such conclusions are warranted by our investigations is to lie. We can help Jim a lot—and we will help Jim a lot more if we stop “diagnosing” him and simply start helping him. That should be our rallying cry: “Lots of help and no more diagnosing!”
A doctor is not engaged in idle investigating. He is trying to succeed in his investigations. We do not think that a doctor has been successful if he engages in one surgery after another to find out “what is wrong with us.” In that unfortunate set of circumstances, he has not reached a conclusion yet, and so he can’t make a diagnosis. If there ever was a way to “diagnose” in human affairs—and there never will be—we would need to set the bar exactly that high: we would need to be successful in our investigations, and we would need to be able to say, “This is clearly causing that.” That time can never come.
A diagnosis is a conclusion about cause and effect. “You need new spark plugs” is a conclusion about cause and effect. “You say you are anxious, so I will say that you are anxious” is not a conclusion about cause and effect. It is time for society, in the form of its legislators and watchdogs, to end this travesty. Millions upon millions of adults and children are receiving “diagnoses” that make no earthly sense. And these “diagnoses” stay with them forever. Mention that you are sad to the wrong person and you will carry a “clinical depression” label with you everywhere.
It is time we placed a moratorium on this illegitimate “diagnosing.” No new system will prove legitimate because we do not actually know what “causes” individual human experiences like sadness and anxiety. It is simply improper to turn human experiences, even of the most painful and unwanted sort, into “disorders.” Let us help with the pain; let us really help Jim. And let us leave “diagnosing” to car mechanics and their faulty carburetors and to medical doctors and their Anton’s syndromes.