Is the diagnosis of neurocognitive disorders actuarial or clinical judgment based?

[neuropsychstudent2021]

I was reading a paper 10.3233/JAD-200778 about actuarial criteria for diagnosis of major and mild NCD. This criteria seems to have strict cut offs and doesn't take into account decline from a previous level (unlike the dsm criteria).

Is this how it's diagnosed in clinical practice? I was under the impression that the diagnosis is a clinical opinion based on multiple data points and not a clear cut yes/no.

---

Members don't see this ad.

 

[AcronymAllergy]

You could make the argument that by requiring presence of "impaired" scores (i.e., >1 SD below the normative mean), the authors are implying that decline is present, even if not explicitly identified.

Although I don't know any neuropsychologists in clinical practice who diagnose mild or major NCD based solely on an actuarial method such as used in the article. I do know plenty of folks who use/recommend such methods as one means of informing clinical judgment. In some ways, it's just a more formalized method of what neuropsychologists have for decades referred to as profile analysis.

Edit: As a related aside, I saw Mark Bondi present on what I imagine was some of this data at INS a few years back (i.e., pre-COVID). It was a great talk.

 

[PsyDr]

Any diagnostic criteria that is not DSM or ICD is clinically useless. You can’t bill for actuarial diagnostic stuff. In a forensic or malpractice case, you can’t defend some departure from the community standard of care because some academic people wanted to change the definition.

You could make the argument that by requiring presence of "impaired" scores (i.e., >1 SD below the normative mean), the authors are implying that decline is present, even if not explicitly identified.

DSM5 criteria is 1 SD below demographically adjusted norms.

 

[AcronymAllergy]

Any diagnostic criteria that is not DSM or ICD is clinically useless. You can’t bill for actuarial stuff. In a forensic or malpractice case, you can’t defend some departure from the community standard of case because some academic people wanted to change the definition.


DSM5 criteria is 1 SD below demographically adjusted norms.

Yep, forgot to mention that; DSM-5 specifies that for mild NCD, "performance typically lies in the 1-2 standard deviation range." Although I wish they'd listed it in the diagnostic criteria as an operationalization of "modest impairment in cognitive performance," rather than semi-hiding it in the "Diagnostic Features" section.

 

[neuropsychstudent2021]

You could make the argument that by requiring presence of "impaired" scores (i.e., >1 SD below the normative mean), the authors are implying that decline is present, even if not explicitly identified.

Although I don't know any neuropsychologists in clinical practice who diagnose mild or major NCD based solely on an actuarial method such as used in the article. I do know plenty of folks who use/recommend such methods as one means of informing clinical judgment. In some ways, it's just a more formalized method of what neuropsychologists have for decades referred to as profile analysis.

Edit: As a related aside, I saw Mark Bondi present on what I imagine was some of this data at INS a few years back (i.e., pre-COVID). It was a great talk.

But wouldn't some people have lifelong impairment in some domains? I believe in Petersens original paper on MCI that it was critical to exclude those people.

 

[AcronymAllergy]

But wouldn't some people have lifelong impairment in some domains? I believe in Petersens original paper on MCI that it was critical to exclude those people.

Yes, they could, which is one reason why few, if any, neuropsychologists would ever only use an actuarial method based solely on test scores to diagnose.

 

[PsyDr]

But wouldn't some people have lifelong impairment in some domains? I believe in Petersens original paper on MCI that it was critical to exclude those people.

Because the word “impairment” refers to a decline relative to baseline. If you ignore that definition, I have a lifelong impairment in my pro basketball abilities. And not being a jerk.

 

[WisNeuro]

To be clear, actuarial methods does not mean you do not take relevant patient history into context. No competent neuropsychologist is skipping gathering history and demographic info and just going right into testing and diagnosis.

 

[WisNeuro]

Because the word “impairment” refers to a decline relative to baseline. If you ignore that definition, I have a lifelong impairment in my pro basketball abilities. And not being a jerk.

I feel like it's a bit of nature and nurture. I mean, I'm sure a little of it was inherent, but I kind of had to work on it a little bit to reach the level of ahole that I was really comfortable with.

 

[PsyDr]

I feel like it's a bit of nature and nurture. I mean, I'm sure a little of it was inherent, but I kind of had to work on it a little bit to reach the level of ahole that I was really comfortable with.

Granny looks you in the eye and asks, “How long until I don’t recognize my husband?”.

Some will wuss out and defer to neurology.

Some will wuss out, sell false hope and say, “we don’t know and science might come up with a cure!”.

Some will hold her hand, cut the BS, tell the truth, and go home to deal with their own emotional responses.

 

[psych.meout]

To be clear, actuarial methods does not mean you do not take relevant patient history into context. No competent neuropsychologist is skipping gathering history and demographic info and just going right into testing and diagnosis.

But in his paper on this, doesn't Meehl talk about actuarial methods as including that history, as well as observations and other qualitative data? Like, your actuarial table would include sections for family history of neurocognitive disorders, mental health Dx, etc.?

 

[neuropsychstudent2021]

To be clear, actuarial methods does not mean you do not take relevant patient history into context. No competent neuropsychologist is skipping gathering history and demographic info and just going right into testing and diagnosis.

Right so the actual diagnosis itself is a clinical opinion? It's up to the clinician to decide if the patient has the disorder based on a combination of history, clinical interview etc?

 

[AcronymAllergy]

Right so the actual diagnosis itself is a clinical opinion? It's up to the clinician to decide if the patient has the disorder based on a combination of history, clinical interview etc?

Yes, all diagnoses in DSM-5 (neurocognitive disorders included) are ultimately made according to clinical judgment.

 

[WisNeuro]

But in his paper on this, doesn't Meehl talk about actuarial methods as including that history, as well as observations and other qualitative data? Like, your actuarial table would include sections for family history of neurocognitive disorders, mental health Dx, etc.?

I believe so, I haven't re-read this one in some time.

Right so the actual diagnosis itself is a clinical opinion? It's up to the clinician to decide if the patient has the disorder based on a combination of history, clinical interview etc?

Yes, clinical opinion is different than clinical judgment.

 

[WisNeuro]

Granny looks you in the eye and asks, “How long until I don’t recognize my husband?”.

Some will wuss out and defer to neurology.

Some will wuss out, sell false hope and say, “we don’t know and science might come up with a cure!”.

Some will hold her hand, cut the BS, tell the truth, and go home to deal with their own emotional responses.

This was always my one of my biggest components of supervision. Drilling into my trainees that you always have honest and frank conversations with the patient and family. Other providers for this patient will punt to someone else, or give them false hope, but you owe them the dignity of being honest.

 

[neuropsychstudent2021]

I believe so, I haven't re-read this one in some time.



Yes, clinical opinion is different than clinical judgment.

Sorry if this is a poor question but what is the difference?

 

[WisNeuro]

Sorry if this is a poor question but what is the difference?

It may be more accurate to describe it as clinical reasoning vs. judgment. One is the process vs. a discrete decision.

 

[PsyDr]

It may be more accurate to describe it as clinical reasoning vs. judgment. One is the process vs. a discrete decision.

"Looking at this guy, walking in the door, I think he's faking it."

Vs

"The clinical presentation and the test data are most consistent with malingered NCD."

 

[psych.meout]

Here's the Meehl paper for those who haven't read it or anyone who wanted a refresher

https://meehl.umn.edu/sites/meehl.umn.edu/files/files/138cstixdawesfaustmeehl.pdf

 

[AcronymAllergy]

This was always my one of my biggest components of supervision. Drilling into my trainees that you always have honest and frank conversations with the patient and family. Other providers for this patient will punt to someone else, or give them false hope, but you owe them the dignity of being honest.

Agreed. It can be particularly tough (and is why you have to be attentive during feedback) when the patient seems to have understood what you've said, but then expresses a summary that's inconsistent or not fully consistent with what you've said. So for example, if the patient has mild NCD and you've explained to them the difference between that and dementia, the etiology that you suspect is progressive, etc., and they then respond with, "ok, well at least I don't have dementia, maybe it'll get better." I always follow those types of statements up with restatement and reclarification.

Feedback is a skill like any other, and patients/families won't always accurately remember your feedback no matter how hard you try (and even when you provide written summaries). But IMO, if you can't sit down and compassionately tell someone you think they have the early stages of a neurodegenerative disorder, and honestly explain what that means for them now and in the future, you have no business practicing clinical neuropsychology.

Same goes for sitting down and having a frank discussion in the context of invalidity. You don't have to be confrontational if that's not your style, but you also shouldn't just entirely avoid the issue. In part because if/when that patient sees your notes, they're gonna be pissed.

 

[psych.meout]

Agreed. It can be particularly tough (and is why you have to be attentive during feedback) when the patient seems to have understood what you've said, but then expresses a summary that's inconsistent or not fully consistent with what you've said. So for example, if the patient has mild NCD and you've explained to them the difference between that and dementia, the etiology that you suspect is progressive, etc., and they then respond with, "ok, well at least I don't have dementia, maybe it'll get better." I always follow those types of statements up with restatement and reclarification.

Feedback is a skill like any other, and patients/families won't always accurately remember your feedback no matter how hard you try (and even when you provide written summaries). But IMO, if you can't sit down and compassionately tell someone you think they have the early stages of a neurodegenerative disorder, and honestly explain what that means for them now and in the future, you have no business practicing clinical neuropsychology.

Same goes for sitting down and having a frank discussion in the context of invalidity. You don't have to be confrontational if that's not your style, but you also shouldn't just entirely avoid the issue. In part because if/when that patient sees your notes, they're gonna be pissed.

I'm not going into neuropsych, but this was definitely one of the most important lessons I gleaned from my neuropsych prac and it paid dividends in my other health-focused practica.