Mental Health Inventory (MHI)
The MHI was developed as part of the National Health Insurance Study (Veit and Ware, 1983) and has been studied extensively in a variety of populations. This instrument provides an assessment of several domains of mental health including anxiety, depression, behavioral control, positive affect, and general distress. The full-length MHI consists of 18 items while the abbreviated version has 5 items. The abbreviated version can be used if time is limited but the full-length version has the advantage of generating subscales. The MHI is one of the components of the MSQLI.
Administration time is approximately 5-10 minutes for the full-length version and 2-3 minutes for the abbreviated version.
The MHI is a structured, self-report questionnaire that the patient can generally complete with little or no intervention from an interviewer. However, patients with visual or upper extremity impairments may need to have the MHI administered as an interview. Interviewers should be trained in basic interviewing skills and in the use of this instrument.
The scoring system for the MHI is relatively complex and generates a total score as well as subscale scores for anxiety, depression, behavior control, and positive affect.
The MHI is easy to administer and provides a quick assessment of both positive and negative facets of mental health, not just psychopathology. The availability of the four subscales, anxiety, depression, behavior control, and positive affect may be useful to investigators interested in testing hypotheses concerning these different areas of function. It is important to keep in mind that the MHI cannot be used to generate a formal psychiatric diagnosis and is dependent upon self-report. Investigators interested in studying emotional aspects of MS in depth would probably want to consider more extensive assessment methods.
The full-length version of the MHI has a Cronbach's alpha of .93 while the short form has an alpha of .82. The MHI has been studied extensively in large populations and comes with considerable evidence for its validity. In the field testing for the MSQLI the MHI showed good convergent and discriminant validity.