Center for Promotion of Child Development through Primary Care

DSM-PC Diagnoses Predict Psychiatric Morbidity One Year Later

BACKGROUND

The AAP published the Diagnostic and Statistical manual for Primary Care (DSM-PC) (Wolraich, Ed., 1996) to help pediatricians identify mental health disorders and newly defined at-risk conditions of lesser severity. These lesser conditions are “developmental variation” (parent is concerned but child appears to be within the normal range) and "problem" (child has some dysfunction and some but not all criteria for a DSM-IV disorder). We developed a computerized parent questionnaire to facilitate making DSM-PC diagnoses as part of the Child Health and Development Interactive System (CHADIS). We validated CHADIS/DSM for diagnosing disorder (Sturner, Morrel, Howard, 2003a) and the new problem and variation categories (Sturner, Morrel, Howard, 2003b) compared to independent concurrent assessments. However, it is not known whether these new categories have utility in predicting later mental health morbidity.

OBJECTIVE

To describe the prediction of mental health status over one year in a group of inner city preschool and school aged children based on DSM-PC categories.

DESIGN / METHODS

This was a prospective cohort study in which caregivers of a convenience sample of inner city children (3-12 years) coming for health supervision visits were administered CHADIS/DSM and repeated one year later (12 mos avg., range 9-19 months).

RESULTS

Total scores used to represent mental health status (1 for each variation, 2 for each problem, 3 for each disorder) show remarkable stability as evidenced by a significant correlation of 0.73. Children with DSM-PC category diagnoses continue to show similar levels of morbidity one year later, e.g., all children without any initial diagnosis were free of disorder (although 14% had a problem), while most children with disorder persisted with disorder and most children with an initial problem diagnosis continued in that category or worsened. The problem category showed its usefulness by documenting the persistent difficulties of the children who no longer had disorder at Time 2 and showed a positive predictive value of 0.67 for significant difficulties of either problem or disorder one year later.

DSM-PC Status T-1
T-2 Problem
T-2 Disorder
T-2 Problem or Disorder
No Diagnosis (N=14)
14%
0%
14%
Variation (N=10)
20%
20%
40%
Problem (N=18)
33%
33%
67%
Disorder (N=16)
19%
63%
82%

 

CONCLUSIONS

This preliminary study provides the first evidence for the predictive utility of the DSM-PC categories. This provides additional justification for pediatricians to use the DSM-PC classification system for triage since children do not tend to “outgrow” the psychiatric morbidity depicted by the DSM-PC categories.

Poster: DSM-PC Diagnoses Are Not "Outgrown" in One Year


Sturner, R., Morrel, T and Howard, BJ. DSM-PC Diagnoses Predict Psychiatric Morbidity One Year Later. Pediatric Academic Society.57: 2711. 2005