Toward diagnostic related groups in child psychiatry: a preliminary study

Private psychiatric hospitalization of children: predictors of length of stay

OBJECTIVE

To quantitatively identify predictors and determinants of length of psychiatric hospital stay for children.

METHOD

Forty-seven demographic, psychosocial stressor, psychopathology, and disposition variables were statistically reviewed as correlates of length of stay in 100 consecutive discharges from a child psychiatry inpatient service (age range 4-12) in a private hospital. Those with strong statistical significance were then analyzed by multiple regression.

RESULTS

Greater severity of psychopathology (measured by the Children's Global Assessment Scale), greater severity of psychosocial stressors (by Axis IV scale), diagnosis of post-traumatic stress disorder, special educational and out-of-home dispositions, and severe tantrums in hospital all strongly predicted longer hospital stay. Diagnosis of adjustment disorder predicted shorter stay. Together these variables explained 57% of the total variance in length of stay.

CONCLUSIONS

The most powerful of these predictor variables could potentially be measured at the time of admission, thus permitting accurate prediction of length of stay. A set of models was generated for this purpose.

Predictors of length of stay of psychiatric adolescent inpatients

This study reviewed 158 consecutive admissions to an inpatient adolescent psychiatric service for factors that determine length of hospitalization. Variables associated with length of stay included diagnostic category and the treatment variables of number and type of psychotropic medications prescribed. Patients with disruptive behavior disorders had the shortest hospitalizations, those with anxiety and affective disorders had stays of intermediate length, and patients with psychotic/organic disorders had the longest hospitalizations. Treatment with lithium, neuroleptics, or tricyclic antidepressants was associated with longer stays. Demographic variables and variables describing the social and family situation or level of dangerousness were not significantly associated with length of hospitalization.

Reliability of psychiatric diagnosis in referred adolescents

The authors report a reliability study of the DSM-III diagnostic system in a series of 393 adolescent referrals diagnosed by two clinicians over a period of 3 years. Overall reliability (kappa) was 0.59. However, most of the frequently used diagnoses had reliabilities of 0.7 or higher. Dysthymic, Adjustment, Overanxious and Avoidant Disorders showed poor diagnostic agreement (kappa below 0.4). The patterns of agreement across this and other studies suggest that diagnostic criteria for Affective, Anxiety and Adjustment Disorders are in need of further improvement.

Identification and diagnostic issues in special education: a status report for child psychiatrists

Education of children with learning, behavioral, or emotional disorders in the public schools has been rather remarkably transformed over the past decade by implementation of Public Law 94-142, Education of the Handicapped Act. Although children with learning disabilities have been served in far greater numbers than before, children with other disorders, including serious emotional disturbance, have either declined in numbers served or are being served in numbers far smaller than current prevalence rates would suggest are adequate. This paper reviews these trends, along with diagnostic criteria used to determine special education eligibility for such children, and discusses selected interdisciplinary and related trends.

DSM-III-R: too much too soon?

Some of the more important changes carried out in the recently published revision of the DSM-III classification (DSM-III-R) are described, and an assessment is made on whether a revision of this nature was necessary and whether it achieved its aims. Since a large number of disorders have been modified in one way or another and since empirical evidence for many of the changes is lacking, the author concludes that DSM-III-R, rather than illuminating the field, is likely to cause confusion and frustration in the minds of clinicians, researchers and administrators, and that in fact DSM-III-R is a new classification, a DSM-IV in disguise.

Validity of Axis V of DSM-III and other measures of adaptive functioning

In order to address validity issues which have not been adequately explored in the existing literature on Axis V, the authors compared ratings on this axis with ratings on three other measures of adaptive functioning (Social Competence, premorbid and present) in a diagnostically heterogeneous group of 195 referred adolescents. It was found that correlations amongst the measures were moderate, except between premorbid and present functioning, which was high. Axis V was found to be comparable with a measure of premorbid functioning while Social Competence was similar to present functioning.