C. James Klett

NOSIE-30: a treatment-sensitive ward behavior scale.

The NOSIE-30 is a highly sensitive ward behavior rating scale. Final item selection includes the best 30 of an original pool of 100 items, tested on a sample of 630 chronic schizophrenic patients. Both status (pre-treatment) and change (post minus pre-treatment) item responses were factor analyzed. Six factor scores are obtained: Social Competence, Social Interest, Personal Neatness, Irritability, Manifest Psychosis, and Retardation. Cross-validation of these six factor scores and a composite score, Total Patient Assets, was obtained in an independent, double-blind drug study conducted by the NIMH-Psychopharmacology Service Center. The NOSIE-30, especially designed for use by subprofessional personnel, provides an economical and accurate means of systematically assessing patient status and change.

Abnormal symptoms, signs, and laboratory tests during treatment with phenothiazine derivatives

Complications were neither frequent nor severe in 599 newly admitted schizophrenie patients treated for 12 weeks with chlorpromazine, triflupromazine, mepazine, prochlorperazine, perphenazine, and phenobarbital. Twelve patients were dropped from treatment because of adverse symptoms or signs, 5 because of hematologic abnormalities, and 4 because of deviant hepatic tests.

An Empirical Comparison of Factor Scoring Methods

Various procedures have been used to estimate factor scores and the present study empirically compared the differential effectiveness of factor scoring by unit weights, by fractional weights using the factor loading, and by fractional weights estimated by multiple regression. The results indicated that these methods of scoring factors are roughly equivalent insofar as the intercorrelations among factor scores or reliabilities are concerned. However, estimation of factor scoring weights by multiple regression was the best method for maintaining orthogonality among the factors and appeared more sensitive in detecting group differences when factor scores derived by this method were used as dependent variables.

Ward-Observable Psychotic Behavior Syndromes

hospital ward provides a broader, more representative sampling of the patient’s interpersonal interactions with a variety of other patients, with nurses, ward aides, and with physicians. For this and other reasons, such as economy, ward observations have been recorded with varying degrees of care since the turn of the century. The study reported here represents an effort to further the development of a quantitative polydimensional conceptualization of observable ward behaviors. The major aims of the experiment were (a) to test for the ward symptom patterns identified in an earlier study (Lorr & O’Connor, 1962), and (b) to identify any higher level ward patterns measured by the Psychotic Reaction Profile (PRP) (Lorr, O’Connor & Stafford, 1960).

Restabilization on Methadone After Methadyl Acetate (LAAM) Maintenance

Sixty-eight heroin addicts maintained for 40 weeks on LAAM or methadone in a doub le-blind study were transferred to a uniform dose of 60 mg. methadone daily at the end of their tenure in the study and followed double-blind for the ensuing six weeks during which their daily methadone doses were adjusted according to their clinical needs. Patients were observed for symptoms and signs of discomfort and for the amount of illicit drug use during this period of transition. The results indicate that patients maintained on LAAM can be readily restabilized on methadone and that sudden decrease of methadone dose tends to result in patients’ supplementing with illicit heroin and, conversely, increasing methadone doses resulted in a corresponding reduction in illicit drug use. It is suggested that a chronic covert abstinence syndrome may exist in some patients on long-term methadone maintenance and that while it may contribute to their continued illicit drug use, it may have a different pathophysiological basis and require different therapeutic considerations.

A Study of Reliability Using Canonical Correlation

The present analysis was undertaken for several reasons, not the least of which was the desire to explore the use of canonical correlation with the type of multivariate data usually obtained in treatment evaluations. Canonical correlation is a technique for estimating the maximum correlation between two sets of variables. Applied to problems of reliability or interrater agreement it provides an upper bound estimate (rather than a lower bound) which might be more useful for some purposes. Because it takes into account all of the variables in each set, the resulting coefficient should be a higher level indicator of agreement than the coefficients associated with any of the syndromes taken singly. Finally, the technique provides information about the number of independent but significant ways in which the two sets are related. Since the author has extensively examined the first and second order dimensionality of the IMPS, it was of interest to determine the number of orthogonal linear combinations that would be yielded by this analysis. The 822 patients were a random sample of psychotics from 16 state and university hospitals and have been described elsewhere (Lorr & Klett, 1965). Each patient was interviewed within the first week of admission to the hospital and rated immediately thereafter on the IMPS by the interviewer and an independent observer. The ratings yielded a set of scores for the observer and the interviewer on the following variables: Excitement, Hostile Belligerence, Paranoid Projection, Grandiosity, Percepmal Distortion, Intropunitiveness, Retardation, Disorientation, Motor Disturbances, and Conceptual Disorganization. The two sets of variables were intercorrelated and the canonical correlations computed. It is possible to obtain as many canonical correlations and the associated pairs of linear functions as there are variables in the smaller of the two sets. With the present data therefore, it was possible to find as many as ten significant canonical correlations and this was the number yielded by the analysis. They ranged in descending order from ,966 for the firsc to .739 for the tenth. This implies a high degree of agreement berween two raters. That there were ten independent ways in which the interviewer set of ratings were related to the observer set of ratings provides additional evidence for the existence of ten independenr dimensions in the data. REFERENCES