Dong Won Cho

Suicide rates among public mental health patients

Age‐specific suicide rates are presented, based on 207 white patients of the Missouri Department of Mental Health who were identified as having committed suicide during 1972–74. Results, divided by age, sex, diagnosis and patient status, are compared with other studies. Male inpatients are about five times more likely to commit suicide compared to the general population, while female inpatients are about 10 times more likely to do so. In both sexes, the rate is greatest for the diagnosis of major affective disorder. A history of psychiatric treatment increases the suicide risk more for women than for men, although male patients are still about twice as likely to commit suicide than are female patients. A quantitative model is presented which describes the relative influence of age, sex and diagnosis on suicide rates.

Patients of a Public State Mental Health System Who Commit Suicide

Twelve per cent of suicides in Missouri during a 3-year period were identified as having been patients of the state mental health care system by computer matching of death certificate tapes and a statewide case registry file. The percentage of Missouri suicides who had been patients varied markedly among age groups. Compared to nonpatient suicides, the patient suicides had a more equal male/female ratio and were younger. Diagnostically, schizophrenia, alcoholism (among males), and affective disorder (among females) were prominent. Nonwhites had a low incidence of suicide. The significance of the patient status at the time of death is discussed.

Computer consultation for emotional crises: An expert system for “non-experts”

Abstract This article describes an expert computerized consultation system for assisting a non-mental-health-expert assess and treat emotional and behavioral emergencies in a remote area. It discusses some of the issues and limitations of this system, and suggests that the intended educational/instructional aspects of expert systems may make them more appropriate for, and potentially better accepted by “non-experts.”

Inter-Rater Reliability: Intraclass Correlation Coefficients

A computer program to calculate six different forms of intraclass correlation coefficients is described. The computational formulas used, input preparation and output description are given. The criteria in choosing an appropriate form of the intraclass correlation coefficients are briefly mentioned.A computer program to calculate six different forms of intraclass correlation coefficients is described. The computational formulas used, input preparation and output description are given. The criteria in choosing an appropriate form of the intraclass correlation coefficients are briefly mentioned.

Missouri Actuarial Report System (MARS)

INCE 1966, the Missouri Department of Mental Health (DMH), in collaboration with the University of Missouri, has been developing a comprehensive statewide computerized information system.‘.2 Ten major mental health facilities are connected via remote terminals with high-speed card-readers and line-printers to an IBM 370-155 computer in St. Louis. Approximately fifteen satellite facilities also process data through the terminals of these ten facilities via messenger service and telephone. The system currently processes about 1000 new patients per month, with over 180,000 patients on the data-base. More than 10,000 clinical reports and 30,000 outpatient contacts are processed monthly. Most of the basic clinical forms are automated, including face sheet, mental status checklist, community adjustment scales, psychologic screening tests, and others.” Early in the history of the project, there was considerable interest in the feasibility of classifying individual patients into clinically useful categories, utilizing the large DMH standardized data-base in establishing explicit rules. These rules usually took the form of equations derived by using multivariate statistical techniques. The term “actuarial” in the behavioral science literature has come to mean this process of delineating and using explicit rules to classify individuals or predict their behavior. In this paper, the terms “actuarial rule set” or “rule set” will refer to such rules or equations. The authors’ earliest attempt at actuarial classification involved a list of major diagnostic categories attached to the printout of a computerized mental status report.’ Associated with each category was a probability estimate of the patient’s belonging to that group.” Later, a separate report was developed concerning risks of short versus long length of hospital stay, unauthorized absence from the inpatient service, and harm to self or others.“PX Thereafter, a single report was designed to incorporate all of the above, plus type of psychotropic drug treatment most likely to be prescribed in the Missouri DMH. Entirely new actuarial rule sets were developed, using more sophisticated statistical techniques. After two years of development, the Missouri Actuarial Report System (MARS) became operational. and the creation of a MARS report was “triggered” automatically whenever an admission inpatient mental status report was processed by the computer.

An Iterative Program for Scale Building Based on Intra-Scale Consistency

an initial pool of items. The choice of the initial set of items depends upon the judgment of competent persons as to their suitability (face validity) for the purposes of the scale. The aim is to identify a subset of these items with high internal consistency. The computer program calculates and prints out the mean, variance, and standard deviation for each item in the initial pool. It also accomplishes the following iterative procedure: