NEW STUDY: NPD Cured? NOT!

Weinberg, Igor PhD; Ronningstam, Elsa PhD; Ravichandran, Caitlin PhD; Gunderson, John G. MD. Can Patients With Narcissistic Personality Disorder Change? A Case Series. The Journal of Nervous and Mental Disease 212(7):p 392-397, July 2024. | DOI: 10.1097/NMD.0000000000001777

Ronningstam, Gunderson leading scholars on narcissism. But the study is dramatically and irredeemably flawed. It flatters and fawns on both these eminent scholars and, evidently, they are far from immune to it.

8 patients: impossible to establish statistical significance.

Shockingly: the clinicians working with these patients were the ones who selected and evaluated them! Of course they would say that the treatments have been an astounding success, reflecting well on their own miraculous professionalism. The authors admit that the clinicians selected ONLY patients who have improved, never the patients who haven’t (sampling bias). This means that the results cannot be generalized to the general NPD population.

The researchers supposedly came up with their own diagnoses which mysteriously tallied closely with this catastrophically skewed non-representative sample. Of course, the researchers are even more biased than the clinicians in this case!

“Comorbidities were common (average 2.5/patient). In terms of comorbid disorders, five had mood disorder, including bipolar II disorder (n = 1), substance use disorders (n = 3), eating disorder (n = 3), and OCD (n = 1). In addition, four had trait or diagnosis of borderline personality disorder, and one had antisocial features.”

Diagnostic Interview for Narcissism (DIN) and the Diagnostic Statistic Manual for Psychiatric Disorders, 5th Edition (DSM-5) Personality Disorder Section II categorical criteria outdated.

Nothing new regarding the improvements in psychosocial functioning (behavior modification).

No differentiation between critical clinical features (such as lack of empathy, grandiosity, narcissistic supply/regulatory self-enhancement) and nonessential ones (such as exploitativeness, envy, entitlement). The polythetic problem is one of the reasons the DSM IV text (incorporated in the DSM 5) should never be used in research.

Retroactive pretest-posttest design makes it impossible to tell what is the etiology of any observed changes in the patient: therapy? life circumstances and events? Relationships? attempts to impress or manipulate the clinician? Threats to internal validity (e.g., testing, history, maturation).

How do we know that the changes are the outcomes of therapy and not other factors or confounding factors? You guessed it: the clinicians who have administered said therapies assure us that this is the case!

Majority of the patients took part in multimodal treatments, we do not know what aspect of the treatments reported in the case reports contributed to this outcome.

The very awareness of participating in the study affects the behaviors and responses of narcissists who constantly manage impressions.

Retrospective assessment of NPD at baseline and posttreatment is problematic. It involves recall bias, selection bias, difficulty in establishing causation, missing information, generalizability, confounding variables, temporal relationships, statistical limitations (e.g., incidence rate).

Study not really longitudinal: patients have not been followed up after termination of therapy to evaluate the stability of the alleged remission.

The other convos in this Cycle here