7 ms·

It's odd to see a "study" that says psychiatric diagnoses are not evidence-based which itself doesn't actually use empirical evidence. As far as I can tell fro

by wahern 7y ago

It's odd to see a "study" that says psychiatric diagnoses are not evidence-based which itself doesn't actually use empirical evidence. As far as I can tell from the article, the "study" is basically arguing that because most psychiatric diagnoses use circumstantial evidence--that is, they don't rely on tools like an MRI to quantitatively identify and classify a neurologically-based defect--that it's not evidence based. But that doesn't follow. Most of medicine is like psychiatry, at least in the way that that paper seems to criticize. If a person comes into the ER with the symptoms of a cold and you treat them for a cold, you can't say that the diagnosis of a cold wasn't evidence based simply because the doctor didn't run a test to identify rhinovirus fragments. If it's known that the vast majority of such cases would be rhinovirus, that the vast majority of cases for those with something more severe (and where misdiagnosis would be dire) would present with different symptoms, then treating the patient as-if they have rhinovirus is a legitimately evidence-based, experimentally supported activity. It may not be ideal for each particular patient. The doctor isn't searching for some objective truth about each patient's condition. But so what? We do that because we're trying to maximize population wide outcomes. If we abstained from diagnosing every cold without first identifying viral fragments, many people would not only go undiagnosed but completely untreated, including those with much more severe infections. If it walks like a duck and quacks like a duck, it just might be a duck. And if research says that it is a duck 90% of the time, then making that inference is honestly and truly evidence based, even though it leaves open the possibility of 10% misdiagnosis rate. It's also not surprising that the assessment criteria would differ markedly between different types of ailments. The criteria for assessing a sprained ankle wouldn't look anything like that for a cold. It would be highly context based. We can't infer that it's therefore not evidence based. Psychiatry may or may not be evidence based, but you can't show that by an examination of the DSM. You have to affirmatively show the absence of a connection to evidence. Without doing that, you could only at best make a persuasive argument that the evidence itself sucks, or that the disease models are weak. But, again, a weak disease model doesn't imply lack of evidence. Our existing models of most diseases are going to look incredibly weak 200 years from now; but it wouldn't be reasonable to say that late 20th century medicine was unscientific, non-evidence-based hokum.

shkkmo 7y ago
> As far as I can tell from the article, the "study" is basically arguing that because most psychiatric diagnoses use circumstantial evidence Perhaps you would have been better served by reading the linked abstract of the paper. The paper specifically criticizes the heterogeneous and inconsistent of classification of identical symptoms in the the diagnostic criteria for similar diagnosis. >> Selected chapters of the DSM-5 were thematically analysed: schizophrenia spectrum and other psychotic disorders; bipolar and related disorders; depressive disorders; anxiety disorders; and trauma- and stressor-related disorders. Themes identified heterogeneity in specific diagnostic criteria, including symptom comparators, duration of difficulties, indicators of severity, and perspective used to assess difficulties. > If a person comes into the ER with the symptoms of a cold and you treat them for a cold, you can't say that the diagnosis of a cold wasn't evidence based simply because the doctor didn't run a test to identify rhinovirus fragments. You picked a particularly odd example. There is no treatment for colds, all that can be done is to alleviate the patients symptoms. Indeed, there a number of situations in which you would indeed want to confirm that the symptoms are caused by the common cold because there are other potential causes that can be life threatening. [0] [0] https://www.health.com/cold-flu-sinus/signs-more-serious-common-cold?slide=418194#418194 https://www.health.com/cold-flu-sinus/signs-more-serious-com... > And if research says that it is a duck 90% of the time, then making that inference is honestly and truly evidence based, even though it leaves open the possibility of 10% misdiagnosis rate. ... what? A 10% misdiagnosis rate because other possible explanation for symptoms are ignored is considered good science and medicine? > It's also not surprising that the assessment criteria would differ markedly between different types of ailments. The criteria for assessing a sprained ankle wouldn't look anything like that for a cold. Except the study explicitly compares similar ailments (more like "sprained ankle" vs "broken ankle" vs "torn ligament in ankle" The main point of the criticism is that unlike other medical diagnosis, psyciactric diagnosis minimize the role of trauma. This is like a doctor trying to diagnose ankle pain while not really considering the fact that you fell off your bike. > Pragmatic criteria and difficulties that recur across multiple diagnostic categories offer flexibility for the clinician, but undermine the model of discrete categories of disorder. This nevertheless has implications for the way cause is conceptualised, such as implying that trauma affects only a limited number of diagnoses despite increasing evidence to the contrary.

by wahern 7y ago

shkkmo 7y ago

> As far as I can tell from the article, the "study" is basically arguing that because most psychiatric diagnoses use circumstantial evidence Perhaps you would have been better served by reading the linked abstract of the paper. The paper specifically criticizes the heterogeneous and inconsistent of classification of identical symptoms in the the diagnostic criteria for similar diagnosis. >> Selected chapters of the DSM-5 were thematically analysed: schizophrenia spectrum and other psychotic disorders; bipolar and related disorders; depressive disorders; anxiety disorders; and trauma- and stressor-related disorders. Themes identified heterogeneity in specific diagnostic criteria, including symptom comparators, duration of difficulties, indicators of severity, and perspective used to assess difficulties. > If a person comes into the ER with the symptoms of a cold and you treat them for a cold, you can't say that the diagnosis of a cold wasn't evidence based simply because the doctor didn't run a test to identify rhinovirus fragments. You picked a particularly odd example. There is no treatment for colds, all that can be done is to alleviate the patients symptoms. Indeed, there a number of situations in which you would indeed want to confirm that the symptoms are caused by the common cold because there are other potential causes that can be life threatening. [0] [0] https://www.health.com/cold-flu-sinus/signs-more-serious-common-cold?slide=418194#418194 https://www.health.com/cold-flu-sinus/signs-more-serious-com... > And if research says that it is a duck 90% of the time, then making that inference is honestly and truly evidence based, even though it leaves open the possibility of 10% misdiagnosis rate. ... what? A 10% misdiagnosis rate because other possible explanation for symptoms are ignored is considered good science and medicine? > It's also not surprising that the assessment criteria would differ markedly between different types of ailments. The criteria for assessing a sprained ankle wouldn't look anything like that for a cold. Except the study explicitly compares similar ailments (more like "sprained ankle" vs "broken ankle" vs "torn ligament in ankle" The main point of the criticism is that unlike other medical diagnosis, psyciactric diagnosis minimize the role of trauma. This is like a doctor trying to diagnose ankle pain while not really considering the fact that you fell off your bike. > Pragmatic criteria and difficulties that recur across multiple diagnostic categories offer flexibility for the clinician, but undermine the model of discrete categories of disorder. This nevertheless has implications for the way cause is conceptualised, such as implying that trauma affects only a limited number of diagnoses despite increasing evidence to the contrary.