Psychotic Disorders and COVID

[DynamicDidactic]

New study

Association of Psychiatric Disorders With Mortality Among Patients With COVID-19

This cohort study examines the incidence of mortality in patients with COVID-19 who had premorbid diagnoses of schizophrenia spectrum, mood, or anxiety diagnoses.

jamanetwork.com

In this cohort study of adults with SARS-CoV-2–positive test results in a large New York medical system, adults with a schizophrenia spectrum disorder diagnosis were associated with an increased risk for mortality, but those with mood and anxiety disorders were not associated with a risk of mortality.

I am curious about the reliability of the diagnoses in the records review (e.g., misdiagnoses) and the overall covariance of general poor health in individuals with psychotic disorders but I thought an interesting finding.

---

Members don't see this ad.

 

[R. Matey]

I just want to know why anyone still uses SAS when R exists.

It is an interesting finding and I think you're right to be suspicious about the reliability of records review. My experience is that it would depend on where the first diagnosis was made and whether the treatment plan differed enough for a provider to bother to change it (e.g. MDD with psychosis vs. schizoaffective disorder).

 

[DynamicDidactic]

I just want to know why anyone still uses SAS when R exists.

old dogs, new tricks and what not.

 

[WisNeuro]

I am curious about the reliability of the diagnoses in the records review (e.g., misdiagnoses) and the overall covariance of general poor health in individuals with psychotic disorders but I thought an interesting finding.

Don't have the study, just read through the abstract. 1) I don't trust pure record review studies. Just look at the garbage that comes out or record review studies and the VA in TBI. Though, that trash will get you in JAMA, but we all know how that profession is trained in research. 2) Is it possible that this is confounded with individuals (usually elderly) who are prone to delirium with systemic infection I've seen a lot of "psychosis" diagnoses never cleared in the problem list, of which the only psychotc symptoms occur in the course of delirium.

 

[Fan_of_Meehl]

Don't have the study, just read through the abstract. 1) I don't trust pure record review studies. Just look at the garbage that comes out or record review studies and the VA in TBI. Though, that trash will get you in JAMA, but we all know how that profession is trained in research. 2) Is it possible that this is confounded with individuals (usually elderly) who are prone to delirium with systemic infection I've seen a lot of "psychosis" diagnoses never cleared in the problem list, of which the only psychotc symptoms occur in the course of delirium.

I chuckled with recognition at 'the garbage that comes out of record review studies and the VA in TBI.' So true.

 

[WisNeuro]

I chuckled with recognition at 'the garbage that comes out of record review studies and the VA in TBI.' So true.

Luckily, it's easy to trash that JAMA citation when plaintiff hacks use it. Such bad methodology. Particularly that mTBI/dementia one. Undergrads know more about research design than those *****s.

 

[AcronymAllergy]

Haven't read the article, so this is a bit of a lazy/sloppy reply, but I suspect it could in part have to do with all the other health factor comorbidities/increased risks in the schizophrenia population, if those weren't covaried (e.g., increased risk for cardiovascular disease, cigarette smoking, obesity, etc.). Plus the shorter life expectancy in general.

 

[futureapppsy2]

Don't have the study, just read through the abstract. 1) I don't trust pure record review studies. Just look at the garbage that comes out or record review studies and the VA in TBI. Though, that trash will get you in JAMA, but we all know how that profession is trained in research. 2) Is it possible that this is confounded with individuals (usually elderly) who are prone to delirium with systemic infection I've seen a lot of "psychosis" diagnoses never cleared in the problem list, of which the only psychotc symptoms occur in the course of delirium.

Maybe I'm being sensitive because I just did a VA chart review study (approved the proofs today, in fact), but I think that chart review, like anything else, can be done well or poorly. In our case, we spent a lot of time on inclusion/exclusion/validation criteria, and the diagnosis that we were examining (ALS) is much less subjective than, say, mTBI.

 

[WisNeuro]

Maybe I'm being sensitive because I just did a VA chart review study (approved the proofs today, in fact), but I think that chart review, like anything else, can be done well or poorly. In our case, we spent a lot of time on inclusion/exclusion/validation criteria, and the diagnosis that we were examining (ALS) is much less subjective than, say, mTBI.

It can definitely be done more/less rigorous. I've published a study using chart review, though we also had a neurpsych eval, and access to the entirety of the chart review vs the problem list. The studies I'm referencing almost solely got their diagnoses from problem list and made a lot of wild conclusions based on that problematic piece of info.

 

[cara susanna]

Ooh, yeah, I didn't read the article but the problem list is definitely a problematic (haha) way of determining diagnosis.

 

[futureapppsy2]

It can definitely be done more/less rigorous. I've published a study using chart review, though we also had a neurpsych eval, and access to the entirety of the chart review vs the problem list. The studies I'm referencing almost solely got their diagnoses from problem list and made a lot of wild conclusions based on that problematic piece of info.

Oh yeah, looking at the problem list alone would be super problematic. We cross-referenced dx with prescriptions, treatment dates, etc., to further verify.

 

[Fan_of_Meehl]

Ooh, yeah, I didn't read the article but the problem list is definitely a problematic (haha) way of determining diagnosis.

For some charts/patients, a randomly-determined selection of, say, 5 - 20 diagnoses from the ICD or DSM systems would be just about as meaningful and useful. Garbage in, garbage out. But the biggest issue with most of the research on mTBI coming from the VA is that it's always correlational (and they generally fail to meaningfully account for important covariates) yet it is almost always presented as evidence for a causal model of {history of concussion / mTBI --- [causes] ---> clinical issue [e.g., memory complaints, somatic complaints, mental health complaints]).