Can the M.I.N.I. International Neuropsychiatric Interview Be sELf-Administered?

Why the MINI should be clinician-rated

Need for clinician input and expertise

While the MINI is based on patient responses, many of the questions require clinician input. Consider question A3e on Module A for Major Depressive Episode (MDE). Referring to any 2-week period in which a patient reports having felt depressed, A3e asks:

“Did you feel worthless or guilty almost every day?”

Let’s say the patient says “yes.” The directions in small caps direct the interviewer to probe for examples to see if they are “consistent with a delusional idea”:

IF YES, ASK FOR EXAMPLES. LOOK FOR DELUSIONS OF FAILURE, OF INADEQUACY, OF RUIN OR OF GUILT, OR OF NEEDING PUNISHMENT OR DELUSIONS OF DISEASE OR DEATH OR NIHILISTIC OR SOMATIC DELUSIONS. THE EXAMPLES ARE CONSISTENT WITH A DELUSIONAL IDEA.

This is not an assessment that patients can typically make on their own. Nor is it one that doesn’t matter. Failure to identify a  delusion in A3e not only matters to accurate identification of the immediate disorder, MDE. It also matters to accurate identification of other disorders. That’s because A3e is revisited on the Mood Disorders Algorithms page — where it is used to differentiate Major Depressive Disorder(MDD) with Psychotic Features from MDD without Psychotic Features. 

Question C3a in Module C for Manic and Hypomanic Episodes similarly directs the interviewer to probe for examples, in this case of feeling “that you could do things others couldn’t do” or that you were “an especially important person” to see if they rise to the level of a delusion. Again, the distinction isn’t one that patients can typically make independently. It doesn’t just apply to the immediate disorder, manic or hypomanic episode. As in A3e, C3a is revisited on the  Mood Disorders Algorithms page to distinguish between Bipolar Disorder I with Psychotic Features and Bipolar I Disorder without Psychotic Features.

 These are just two illustrations. Multiple questions in different modules of the MINI require clinician input and expertise to see if the examples meet DSM criteria for diagnosis of specific disorders.

Need for clinician observation and judgment

Other questions on the MINI, especially on Module K for Psychotic Disorders, explicitly ask for clinician observation and judgment. Consider, for example:

K18b:

IS THE PATIENT CURRENTLY EXHIBITING INCOHERENCE, DISORGANIZED OR DERAILED SPEECH, OR MARKED LOOSENING OF ASSOCIATIONS?

K10 a:

DID THE PATIENT EVER IN THE PAST HAVE NEGATIVE SYMPTOMS OF SCHIZOPHRENIA, E.G. SIGNIFICANT REDUCTION OF EMOTIONAL EXPRESSION OR AFFECTIVE FLATTENING, POVERTY OF SPEECH (ALOGIA) OR AN INABILITY TO INITIATE OR PERSIST IN GOAL-DIRECTED ACTIVITIES (AVOLITION)?

K10 b:

ARE NEGATIVE SYMPTOMS OF SCHIZOPHRENIA, E.G. SIGNIFICANT REDUCTION OF EMOTIONAL EXPRESSION OR AFFECTIVE FLATTENING, POVERTY OF SPEECH (ALOGIA) OR AN INABILITYTO INITIATE OR PERSIST IN GOAL-DIRECTED ACTIVITIES (AVOLITION), PROMINENT DURING THE INTERVIEW?

The problem isn’t just that terms like “loosening of associations,” “alogia,” “avolition, or “negative symptoms of schizophrenia,” are unfamiliar to patients. There is no way that they can accurately rate items like these on their own since they require external observation by a trained clinician.

Need for Insight

Additional questions on the MINI require insight that patients, because of their underlying disorders, do not possess. I cannot tell you how many times I have interviewed patients who exhibit marked psychomotor retardation in the interview but deny that they “talk or move more slowly than normal “ or that others notice this. (A3c, MDE Module). That’s because they lack the insight, because of their depressive symptoms to see themselves as others do.

Four risks of self-report or patient-rating of the MINI

Risk 1: Inaccuracy and misdiagnosis

In the mid 1990s in our original validation study of the MINI we tested the usability of patient ratings in a subsample of patients. The results were disappointing. While the standard clinician-rated MINI performed well against the two gold-standard interviews —Structured Clinical Interview for DSM Disorders (SCID) and Composite International Diagnostic Interview (CIDI) used for comparison —  this was not the case when the MINI was patient-rated. In fact, the patient ratings performed very poorly — missing diagnoses or producing inaccurate ones — especially for critical disorders such as Psychotic Disorder, Mania, Drug Dependence and patients with multiple disorders or comorbidities.[1] I have not seen any new studies of a patient rated MINI to contradict these results. The FDA defines the MINI as a ClinRo or Clinician Reported Outcome Measure and the FDA Guidance for clinical trials is for the MINI to be clinician rated.  The EMA 2025 guidance documents MINI as a suitable diagnostic instrument since it was validated under conditions in which it was administered in paper format as a primary source document. The results of the validation studies in paper cannot be generalized to any other conditions or format or used as evidence of validation for the FDA or EMA or another regulatory authority if the MINI was not administered on paper as a primary source document.  I am not aware of any validation studies of the MINI that have been done in electronic form and recommend that you get documentation of any validation in electronic form in writing by any third party who sells you software marketed as “validated” as well as documentation that any software being used in diagnosis (1) has been cleared by the FDA with 510K documentation and the EMA or (2) official letter from FDA and EMA  that their software does not diagnose and thus does not meet definition of a device under section 201(h) of the FD&C Act.  Please be aware that under the FDA 2024 guidance regarding CRO and Sponsor liability the Sponsor and CRO are “regulated entities”.

Risk 2: Safety issues

Related to the risk of inaccuracy is safety and the potential for patient harm. Patient-rating of the MINI can lead to a failure to anticipate a suicide, homicide, or physical assault on another person, incidents that have medico-legal and public health consequences. There are many other safety risks:

• What if someone completes the MINI on their own without clinician input, mistakenly decides they have schizophrenia or another serious psychiatric diagnosis, and makes a suicide attempt because of their own self-misdiagnosis?
• What if a diagnosis of current mania is missed and the patient is put on an antidepressant medication that increases their mania and leads to suicide or assaultive behavior towards others? What if the patient is just put on a treatment that is ineffective because of the misdiagnosis?
• What if a patient with current drug dependence is missed and the patient dies from an overdose on a prescribed treatment for another disorder because of a drug-drug interaction?
• What if an employer lets job applicants self-assess using the MINI and through misapplication, some applicants are denied a job for which they are eligible, or they are denied insurance for themselves and their family?

The list is endless. The safety and legal risks of self-report or patient-rating the MINI are not worth it in my judgment.

Risk 3. Threats to data Integrity in trials

The MINI is often used to assess eligibility for inclusion or exclusion in clinical trials of new medications for psychiatric and other medical disorders. Allowing patients to self-rate the MINI could derail a trial from the outset.

• What if the target disorder is missed? Patients who might have been eligible will be excluded limiting the available population to study.
• More critically, what if an excluded disorder such as mania or psychotic disorder is not identified and patients with these disorders are inadvertently included? It may not be possible to demonstrate treatment benefit. What’s more, reportable AEs such as “treatment-emergent mania” or “treatment-emergent psychotic symptoms” could color the treatment’s safety profile.

Risk 4. Regulatory issues

The MINI is recognized by U.S. and international regulatory bodies as a Clinician-Reported Outcome Assessment (Clin-RO) instrument, not as a Patient-Reported Outcome Assessment (Clin-PRO) instrument. Agencies such as the U.S. Food and Drug Administration (FDA), European Medicines Agency (EMA), and other international regulatory bodies require Clin-PRO assessments to be validated, i.e., shown to be fit-for purpose and reliable, before they are used in clinical trials. The same is true for the Centers for Medicare and Medicaid Services (CMS) and third-party payors when it comes to use of the MINI in primary care or clinical practice.

While the MINI has been extensively validated in its standard clinician-rated (Clin-RO) format, it has not been validated in patient-rated (Clin-PRO) format. This is a significant limitation both for trials and for practice. In the case of trials, approval could be denied if the MINI isn’t used in its validated clinician-rated format. In the case of clinical practice, payment could be disallowed.

What about a 2-step operation of patient ratings followed by clinician review? This is not a practice I endorse since the clinician review in such a 2-step process, if done properly, could take as long or longer as the clinician rating in the first place.

Bottom Line

1. The MINI was designed as a clinician-rated diagnostic interview.
2. To be accurate it requires clinician input, expertise, and judgment.
3. Patient-rating of the MINI should be avoided because of the potential for inaccuracy (misdiagnosis), patient safety risks, threats to data integrity, and regulatory reasons.

[1] Sheehan DV, Lecrubier Y, Sheehan KH, et.al. The Mini-International Neuropsychiatric Interview (M.I.N.I.): the development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. J Clin Psychiatry. 1998;59 Suppl 20:22-33;quiz 34-57. PMID: 9881538.

For more information on the MINI review our FAQs

Disclaimer: This article is for educational purposes and does not constitute medical, legal or regulatory advice. Always consult with regulatory experts before implementing the MINI in clinical trials or digital health applications.

To license the MINI or MINI-Kid directly from the copyright holder, consult with Dr. Sheehan or obtain training on the use of the MINI, go to DASH Neuroscience.